My friend you’re on the journey and it’s an amazing skill set to learn. And you’re asking what feels like a simple question. In reality it’s a little more complex. I think we’d all really encourage you to spend some time reading and watching and learning. Another approach is to build a “lab” set up where failure has very low consequence. You’ll get this stuff. Give it time.
#installation-archived
1 messages · Page 39 of 1
I mean, it's not like I haven't been reading and watching and learning. I'm essentially just trying to get HA to work with the reolink doorbell I have...everything else is homekit compatible. I had everything working in a VirtualBox VM running on MacOS Big Sur, but if I left it unattended for longer than 2 minutes, it would freeze, and I'd have to force quit and reboot VB and restart the VM. ProxMox running HAOS was recommended to me as perhaps a more stable setup, so I'm trying to figure it out. But if I can't get it to work reliably within a week, I'm going to return the reolink and go with something that is completely homekit compatible...while the reolink is still within its return window.
I am a caregiver with decent technical skills, but I'm not an IT person by trade or background.
Hey guys, so I'm stuck with my first installation of Home Assistant. I'm running it on an intel nuk using proxmox and podman.
I can't seem to access the web interface of neither home assistant or podman. Let me know if you need anymore information.
I am able to access to proxmox webinterface
I understand that - I know what it’s like to get in deep and get in quickly. I’m not suggesting you need to become an sysadmin overnight. I’m worried you’re trying to tackle multiple problems at once: dns, proxmox, haos, and an integration for reolink. How can we simplify this? Also we have to focus - while many of us also use proxmox or VMs, that’s not the scope of this server. I’m happy to try and help but it’s “out of band”
We need all the information. IPs, how’d you configure everything? Network topology? What’s your client and its IP?
I feel like sharing IP's is a bad idea.
I do appreciate the help, btw. I don't want to come across as ungrateful.
Correct me if I am wrong, please: I can have HA disconnected from the internet, because I can add it to Apple Home as a hub, and therefore Apple Home will provide all of my remote capabilities.
Not at all. We’ve all been there. Just trying to help you scope
That’s directionally accurate. If you want to do anything outside your home you’ll need an apple home hub like an AppleTV or HomePod. And your theory only works for things HA can expose to HomeKit (lights, fans, switches, some sensors…)
So far, that's been the main advantage of HA for my setup...is the ability to expose things like the reolink video doorbell to homekit.
And if that’s your focus, can I also suggest #ios_and_mac-archived - slower conversation but more specific.
Yep, same. I do most of my manual interactions via HomeKit stuff exposed through HA
Ok give me a second:
IPs:
Gateway: 10.0.0.1 (same with dns)
pv0 and pv1 run on 2 seperate nuks connetcted directly to the router on the same vlan.
pv0: 10.0.0.10
pv1: 10.0.0.20
the web interface at 10.0.0.10:8006 is accessible, and through the web shell script I ran the https://tteck.github.io/Proxmox/ script: bash -c "$(wget -qLO - https://github.com/tteck/Proxmox/raw/main/ct/podman-homeassistant.sh)"
it installed successfuly and I can see it on my prox mox. I can see it on my router that it has ip 10.0.0.11, however 10.0.0.11:8123 doesn't resolve.
Screenshot tsunami incoming.
What’s .11? A vm?
yea
Just make a new vm and download haos and install it. Don’t use that script.
I can do that. I guess I thought here because when I had HA working through VirtualBox, it was working perfectly. I used Scrypted to get the reolink opened to HomeKit, and as HAOS was running, everything was just working perfectly. If I stopped intereacting with it directly, though, it would freeze...so I was looking for a way to get just a base home assistant OS running that wouldn't freeze, because then I'd be in a good spot for further exploration.
Not disconnected from the internet, that’s pretty much required for updates, integrations, etc. it doesn’t need to be exposed to the internet though (like accessible from outside your lan)
Just setup haos directly and remove the extra layer with proxmox if you don’t want/need it
I'm installing on a 2013 macbook pro (2.6ghz intel, 16g ram, 1 tb ssd). Is there a way to put HAOS on bare metal without proxmox?
That would be such a waste. I have no experience with apple stuff but if you insist try if this works: https://www.home-assistant.io/installation/generic-x86-64/
The main advantage of apple stuff is that it is what I have. I have this perfectly good MBP just sitting around on a shelf with more than enough power to be a HA and media servier.
Then leave proxmox on it lol
I might have been unclear. The waste comment was because only running HAOS on it rather than proxmox, with which you can do a whole lot more, would be a waste of resources
So you are a proponent of proxmox with a HAOS in a VM?
Yep
cool I figuired it out. The script ran the instal silently, and i was inpatient, so after ~5 min I restarted the server and that killed the install. after reinstalling and waiting for a bit longer, I was able to do it the original way.
Okay. So I will continue following this tutorial, and hope I don't somehow expose myself to the world. 😂
Okay. This is probably a dumb question, but it's where I'm at in the process: "Set a bridge." I am using a thunderbolt 2 to gigabit ethernet connection. Is that what this is talking about? I think it's something like "eth0", but I'm not sure how to verify this.
No clue about your environment but firewalls are your friend
where on the sd card will i find the option to manually change the ethernet IPv4 setting from disable to "auto " ? i changed routers this morning to a router that only has WIFI not ethernet, The new router wouldnt assign a IP to HA so i made a mistake an d set the ethernet IPv4 to disable, now im locked out of HA all together so need to perhaps change that back manually
I'm not sure you can - but you can if you connect a keyboard and monitor
ok thanks
nmcli is your friend
is there anywhere to load kernel modules as well? i´ve conected an mt7662u based usb wifi dongle
In Home Assistant OS? Mostly no
i can see the firmware file in /lib/firmware, but it looks like as if hassOS does not bring the kernel modules
the installation use the standard ovo image, so i guess it´s alpine based as far i can tell
Well, it's BuildRoot based
Looks like https://github.com/home-assistant/operating-system/blob/dev/Documentation/configuration.md is how you'd load modules
That seems to be relatively new, but it's nice to see you can now
Apologies if you've already tried this and rejected it, but might Homebridge be a simpler way to achieve your goal of just exposing one device to HomeKit? That's where I started, and I found it much simpler (while also being much less extensible, of course).
I haven't looked into Homebridge. The combo of HA and Scrypted does the trick quite nicely, if I can get a stable arrangement figured out. I'll look into it, though. Thanks for the thought.
Sounds good! If you're on the right track and not having issues, I don't mean to lead you astray - apologies! It's also true that my experience (and a not-uncommon scenario generally, it seems) is of starting with HomeBridge and eventually getting to a point where it was not able to do everything I wanted to do, which forced me to transition to HA.
It does look like Homebridge has the capability to run as a background service in MacOS, which I'm assuming means it wouldn't require a virtual machine arrangement....
I personally wouldn't try to run it on a Mac unless that machine was set up as a server. It's possible, but I would always want my HomeBridge on an always-on machine with server-like capabilities, just as with HA, personally.
The issues I'm having seem to be with trying to run Home Assistant on top of MacOS through VirtualBox.
I run it on a RPi although I'm putting HA on a new "server" (an old NUC).
Yeah, it's going to be always on and set up as a server. It's my old Macbook Pro, a 2013 model with plenty of resources for such shenanigans, if I can just get HA stable. lol
2.6ghz 4 core intel i7, 16gb ram, 1 tb ssd...
I got HA running through proxmox, but it doesn't seem to stay on when I close the laptop (I know how to make the system stay awake with the display off and the laptop shut in macos, but I'm at the start of the learning curve with proxmox). 😂
Oof, that is frustrating. Sorry I can't offer more help, but I wish you the best of luck!
Thanks. And if all else fails, and I can get homebridge working with the reolink camera and rock solid, that might be a good interim solution while I figure out proxmox. 😂
This looks like it might have the info I need: https://virtualizeeverything.com/2022/11/12/proxmox-on-a-laptop-end-of-2022/
I'm currently running HA in virtual box on a win 10 install. I am thinking of wiping that machine and going full proxmox for vm and containers. I know I can export the VM from Virtual Box for Proxmox, but are there things I should be aware of? passing thru usbs for zigbee and z-wave currently
I'm trying to install HA on Proxmox via tteck's script. I keep running into an issue where the supervisor is unhealthy and the install indefinitely loops. I've tried reinstalling the vm, reinstalling Proxmox, and even reseting the internet router that my NUC is hardwired to. Here's a portion of the log from HA trying to install: https://hastebin.com/share/qoxuyarigi.sql
Any ideas on how to troubleshoot this?
Can you share a screenshot of your VM's Hardware tab and your host/node's Network tab?
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.
Here's the Hardware tab: https://i.imgur.com/9aFoFCC.png
And the Network tab: https://i.imgur.com/3TnDJvt.png
I don't see anything obviously wrong. Would have been too easy. Open the VM's VNC console and run network info
Haha of course it would be - here's the network info: https://i.imgur.com/UABsbyR.png
Idk if it's helpful, but I am able to ping 8.8.8.8 and www.google.com from the VM console as well
Looks fine as well and if you can ping google then DNS should work too. Maybe a good ol' restart does something
I restarted both the HA vm and Proxmox with no luck...
What if you just install it in proxmox the normal way and not with some script?
Isn’t there a vmdk for haos?
Looks like a DNS issue. How are the nics set up in your proxmox?
What is contents of this file? Looks like supervisor updated it
3-07-26 12:09:30 INFO (MainThread) [supervisor.plugins.dns] Updated /etc/resolv.conf
@primal nymph
The HAOS VM is going out to your LAN for DHCP incl. DNS. It likely that's not working properly
I haven't tried that yet because I've successfully used the script before, but might need to. I'm just holding out because the script is just so dang easy..
I'm not sure how to check my NIC setup... could point me in the right direction?
The contents of that file is just this line "nameserver 192.168.1.1"
For the sake of troubleshooting. I have tried updating the DNS in the DNS tab of the pve to 1.1.1.1 as well as back to 192.168.1.1 to see if that did anything - it did not.
You shouldn't need to do anything like that in the VM or proxmox. Is 192.168.1.1 your router and LAN dns server?
What's the network device on your HAOS VM look like?
Is this the first VM you're running on your proxmox?
Yup 192.168.1.1 is my router, but my LAN DNS is set to 1.1.1.1. My network device is shown in this screenshot: https://i.imgur.com/KEJe9nw.png
It is my first VM, but it was working previously. I couldn't get anything to download or update in HA about a week ago, and thought a fresh install would fix the issue. Guess not
Is the gateway on your vm set to 192.168.1.1 ?
it needs to be able to route via gateway to 1.1.1.1 for internet DNS
It looks like it: https://i.imgur.com/S0dNAbv.png
Is that set in the resolv.conf file?
You shouldn't have to do anything in your VM. Should be in proxmox and your router.
that route looks OK
and the cli can resolve and ping google.com etc?
No need to export the vm, just do a full HA backup, then restore it into your new proxmox install
pass through should work when you configure it. If it's not the same bit of hardware, you should try it side-by-side, before you nuke your old install
So, installing proxmox (started the process over for...reasons), it wants me to put in something for "management interface". Right now, it says "pve.example.invalid". I'm following the tutorial linked to me, and it doesn't really spell this out, and searches on the matter aren't making sense to me. What do I put in this field? Thanks.
afaik the last two weeks of data (entity data) is written to db after 2 weeks. Where is the intermediate data stored?
Huh? Data is constantly written to the DB
Are you confusing the fact that by default 10 days of data are kept?
Sorry, 10 days. I recall that for the first 10 days the data is stored elsewhere OR differently, which leads to some data only being accessible to certain frontend cards
I will reboot my brain then.
Strangely, the reason I asked has now fixed itself.
I was not able to see any data in the frontend (e.g. apexcharts card) from the last 7 days. And history only showed my data > 10 days ago.
Now, after another reboot and having the graph open in a background tab, the data is all there.
Sorry for false alarm.
Btw, @humble mirage , I have now switched to container and am making it work without addons (except HACS). More work but let's see how it goes.
Let's see if your advice pays out in the long run 😉
It will, if you invest in learning 
Trying to 🙂
How do you remote edit you /config/ files?
Did you create a fileshare and then remote edit via e.g. VSC?
I always found it quite nice to be able to edit in the HA webUI in VSC when testing.
So carry on using it?
Or you can run VSC locally and use SSH for it to access the files
Sorry I was afk. I don't think it matters, probably used for certs or something.
I think @vague hill is talking about statistics vs recorder
Will check that out. Looks easy enough. Although it was nice to be in HA and edit there directly. 🙂
Is there a post install checklist ?
Other than https://www.home-assistant.io/getting-started/, no
What are common way and tools to manage the configuration of hass, lovelace, automation etc... ?
Personal preference really
Many people entirely manage the frontend in the frontend itself, zero YAML
Some people only use the automation editor, some write YAML, some use external tools
etc
On my legacy install i edit yaml, except for the automations
But I have pretty simple automations yet
I do everything manually, using vi, but I'm old school
I do as much as I can through the ui, only dipping into other files where I can't do it in the ui
Just old.
Yeah the CLI in both the VM and pve node can ping 8.8.8.8 as well as www.google.com.
What ports are required outbound from HA to get updates?
ok going to see what happens when i just allow 443 out as i hand off NTP via DHCP.
is your NTP server internal ?
Hey everyone, I could use some assistance with my Home Assistant installation. It seems to be stuck at the message 'Host information reload completed' during startup. I've waited for a while, but nothing seems to be happening beyond this point. Any ideas on what might be causing this and how I can resolve it? Thanks in advance for your help!
what hardware are you on?
@primal rapids if you have restrictive outbound policy, can't you just monitor your firewall for denies?
what method of install?
HAOS?
yeah on the rpi patience is a virtue
Good morning everyone 🙂
Is there a way to get update notifications in a container installation?
I just read that I have to update via the host rather than from within HA. But a notification would be nice 🙂
you can run anything on the host from the container also, if thats suitable for you thats up to you (security wise)
and yes, create that notification yourself by setting up the appropriate version entities and a automation
Guys i have installed home assistant OS on a laptop, is there any way to turn off the laptop screen and backlight?
maybe you can get some useful information from here https://community.home-assistant.io/t/turn-off-laptop-screen-running-hassio/228697/17
specifically this looks like a good solution which allows you to actually change it in HA https://community.home-assistant.io/t/turn-off-laptop-screen-running-hassio/228697/41
Only downside is that it'll be a 'permanent' change. if you lose network access, it'll be off 'forever' (though I guess you can run the command yourself) that's if you have an automation which runs it on startup.
might be worth doing it with time delay on restart.
Hello everyone i have problem running home assistant on raspberry pi inside docker can i ask my problem here?
ask away,
Don't ask to ask, just ask your question. Then people can answer when they're around.
When you do ask a question, try to provide as much background detail as possible. Ask yourself these questions first so that others don't have to:
- What version of the Home Assistant are you running? (remember, last isn't a version)
- What exactly are you trying to do that won't work?
- Is the problem uniform or erratic?
- What's the exact error message?
- When did it arise?
- What exactly don't you "get"?
- Can you share sample code, ideally with line errors where the error occurs?
i have installed docker version of hass
when i run this command it works fine
sudo docker run -d --name homeassistant --privileged --restart=unless-stopped -e TZ=Asia/Tehran -v /home/pouya/hass:/config -p 8123:8123 ghcr.io/home-assistant/home-assistant:stable
and i can access HASS Ui from my netwrok
but when i switch to host network it does not work any more
sudo docker run -d --name homeassistant --privileged --restart=unless-stopped -e TZ=Asia/Tehran -v /home/pouya/hass:/config --network=host ghcr.io/home-assistant/home-assistant:stable
i can see the port is listening using sudo netstat -tuln | grep 8123
i am on rasbian ( rasbery pi ) and i dont have any fire wall
i can ssh into my raspberry pi
i tried to disable my ipv6 but nothing has changed
Can't you use something like setterm -blank 1? And if it works, run it as service?
And there are no error or log messages when you start the docker? Did you take a look at it via kvm manager gui and check which ip is assigned to the connection?
Hi,
I'm having some trouble that I hope maybe somebody can help me with.
I'm trying to migrate my hassos over to a new server.
Backup and restore isn't working. Have failed no mater what I try.
I'm now trying to transfer the settings and configs over to the new one using ftp.
But what folders do I need to copy over so that the new instance is a duplicate of the old one?
With both network og HA configs? 🙂
For just HA you'd want the entire contents of your config folder, but there's no way to do that for network settings
For a restore you may be better off doing a minimal onboarding and then using Samba to upload the backup
regarding the network setting, it is mostl because I want the duckdns also to work after migration
That'd be add-on settings
what do you mean about minimal onboarding?
Ah, that's what I have done.
and when I have copied the entire config folder from the old to the new system, but after I have done that, I'm not able to connect to HA. Not with the local IP or duckdns address
Well, that's because the config folder is only for HA (Core)
So, nothing about any add-on, nothing about networking, etc
with the fresh install I'm able to do the minimal onboarding on ip ..10.156, but after the transfer I'm not able to connect to that adress any more 🤔
You need to restore a full backup, not copy the config folder over
and that equals copying the folders addons, backup, config, media, share and ssl? that is accessible thru samba?
A full backup probably includes more than you'll get from copying those locations
Don't you have to reinstall the addons manually anyway tho Tink?
🤔
I'll always be able to reinstall the addons, the problem is to be able to log into ha after restoring the backup on the new hardware
Restore a backup, it restores everything in the backup, no need to manually do the add-ons
I seem to remember a backup some years ago that had all my addon configs but I had to reinstall the addons manually
But I'm also old
Anyway I love that I'm wrong on this one
Cuz boy that was annoying then again I think I was still on an rpi1B+ at that point
But how should I go forward with to get a full restore when in inbuilt restore is not working?
When you've restored it should (I believe) use the old IP if you set a static IP before
When I enter the console, the ip is the same.
But when I go to that ip address, nothing happens
Reboot?
have you tried turning it off and back on again?
my pigeons have been bested once again
I really hate this lag
hello dear respected senior I have issue to start with esp home assistant with the very first step
Open page in your Home Assistant?
and provided with this link
Your instance URL: https://homeassistant.local:8123
but it didnt open web ui
did you try http://homeassistant.local:8123/
, yes it gives an error when I click the open link it says This site can’t be reachedCheck if there is a typo in homeassistant.local.
do you know the IP of HA?
no how to know that if you please guide me
well, you can check your router DHCP list, or you can get to the console of HA
how is your HA installed? on what hardware?
i didnt did anything cause i jump on esp home web i saw the get started with my ad on so i click it
and my ip address is 192.168.18.1 i type ipconfig in command and found it
i have esp32 and i want to use esp home to configure it accordingly is there any thing that need to be done before the ad on?
Have you read the docs at https://esphome.io/?
yes i read them
Do you just want to use ESPHome, or do you want to use Home Assistant too?
and i have issue with the url this url Home Assistant URL
http://192.168.31.205:8123
Note: This URL is only stored in your browser.
You don't need Home Assistant to use ESPHome after all
i want both
and my ip address is 192.168.18.1
http://192.168.31.205:8123/
That's two different networks
Do you have two routers?
yea i try different method
Well, everything should be on the one network
no i have one router and let me share my ips
Wireless LAN adapter Wi-Fi:
Connection-specific DNS Suffix . :
IPv6 Address. . . . . . . . . . . : 2400:adcc:10c:4100::3
IPv6 Address. . . . . . . . . . . : 2400:adcc:10c:4100:9bf7:b98f:dd78:f575
Temporary IPv6 Address. . . . . . : 2400:adcc:10c:4100:34d9:b8ad:e186:c2e3
Temporary IPv6 Address. . . . . . : 2400:adcc:10c:4100:aca8:fc91:96fb:509c
Link-local IPv6 Address . . . . . : fe80::d6ad:afc4:e299:312d%15
IPv4 Address. . . . . . . . . . . : 192.168.18.54
Subnet Mask . . . . . . . . . . . : 255.255.255.0
Default Gateway . . . . . . . . . : fe80::1%15
192.168.18.1
HA should also have an IP starting 192.168.18.
If it doesn't then you need to fix that
How did you install Home Assistant?
Is it in a virtual machine? On a Pi? Something else?
i am using laptop and try HA web https://esphome.io/guides/getting_started_hassio.html
Installing ESPHome Dashboard
The ESPHome Dashboard can be installed as a Home Assistant add-on, which you can find in the add-on store in the Supervisor panel. Open it using the following button then click on INSTALL:
If you do not have the official ESPHome add-on repository added you can add with this button:
Open your Home Assistant instance and show the dashboard of the ESPHome add-on.
After that, wait a bit until it is installed (this can take a while). Click “Start” and then click “Open Web UI”. If you see “502: Bad Gateway” it is currently starting, and you can refresh the page after a couple of seconds.
ok
no
You flashed HAOS directly to your laptop?
i am using on laptop window i didnt emulate any virtual machine
So ... where did you install Home Assistant?
no i ddint flash anything yet
i didnt install it yet i am trying to instal on esp32
Ah
You've missed the point of the instructions you linked to
installed as a Home Assistant add-on
You have to have installed Home Assistant before you can use Home Assistant
ok can you prefer a link youtube or anything that help my understanding
The topic of this channel is:
Discussion about how to install Home Assistant (https://www.home-assistant.io/installation), and issues with the install itself. Make sure to mention which type of installation you have. If you're not sure of your installation method, check under Info in the System Health Card (https://my.home-assistant.io/redirect/info/)
There's a link to the Home Assistant install docs in the channel topic
ok
And there, in the bot message
thanks and sorry for messing up i wil follow it thanks it help me alot
No worries, it's easy to miss the woods for the trees
He guys I'm gonna switch from smarthings to home assistant. I read that a Thin client a good replacement is instead of a rpi. Which one do you guys recommended?
Whatever fits in your budget and requirements - but #hardware-archived has a long history of recommendations
You want a CPUMark of over 900 (Pi4 equivalent) - I'd suggest 2000 or better, at least 4 GB of RAM (more if you want to run lots of add-ons), and at least 32 GB of disk space (but again, more is better)
do i have to use etcher to make the install usb from? i extracted the img and wrote that to a usb and when i booted from it it took me directly to the HA CLI 🤔
https://www.ebay.com/itm/185980305924 $80 or https://www.ebay.com/itm/385670237223 similar pricing but options for choosing more ram/ssd size
Is the Dell Wyse 5060 good ennough ?
for the price thin clients arent really worth it in my opinion
There is no "install USB", it's not like a Linux distro or whatever where you make a live USB, boot it, and install to the internal drive. What you did is install HAOS to the USB drive. You want to create a live Linux USB (or make a Ventoy USB, which is an extremely useful tool you should take a look at it), and then etch the target device's internal drive from the Linux you booted.
https://www.home-assistant.io/installation/generic-x86-64
hi folks, i think this is the channel to ask about tech problems? wonder if anyone can give me some tech help. i blew up my previously working HA instance - totally my fault, but i do have a backup so i don't need to recover the data (yet). basically i tried to rearrange the storage SSDs i was using on my ARM board, but messed it up. trouble is, my ARM board is not booting without the SSD available. i'm using the HASS image on an ODROID XU4. it seems to stop at the 'Launching Home-Assistant CLI' screen. no ports are open. is my only option to reimage the device? i'm a Linux professional so if i can fix it without a reinstall, i'm not afraid of the command line. any help appreciated
looking for help on a corrupted HA boot disk on RPI after lightening strike, now it will not boot, I have it on a RPI as a USB mounted card reader, it shows 5 partitions and 2 wont mount, how to I rescue the disk
Are you using Windows?
Windows doesn't know how to read the file systems Linux uses
it is on a RPI as a USB mounted sd card reader
so you've booted the Pi into Linux from its own SD card and are trying to access the HA SD card that way?
yes that is correct
try fsck /dev/sdXN where X is the card reader and N is the partition number that won't mount
if it starts asking questions, ping us back before answering them
You want the data partition
sorry newbie here, can i upload image here to show you of screen shot I have?
looks like you need to upload the screenshot to Imgur first and link it here, not upload it directly
via this
thanks. here are the images
thanks for the link, now at least seems reasonable to have so many partitions, when i inserted disk, it said disk 3 and 5 would not mount due to error
can you help with the exact commands to try to rescue this?
Can you mount hassos-data?
im at the onboarding start page, is it cool to shutdown the machine from here and boot it back up to the same spot when im ready to migrate/restore from backup?
sorry can you be more specific on command line? Like "sudo mount hassos-data"
and if yes, is there a safe way to shutdown from the ha cli 🤔 it doesnt like shutdown
Can you not just clck it in the GUI there?
Otherwise you need to know what the device path is for it
ok I was able to mount it and this si what i see
that is what you should see, so the partition is intact
great so what is the best approach from here? try to rescue it or do a fresh install and inject the data folder?
i would definitely back up the contents of the data partition somewhere safe
The disk took a hit from lightning, don't assume it'll keep working
no idea if shutting down from that state was ok but the command was host shutdown.
if it actually took a hit from lightning, an SD card would be instantly toast. but yes, in this state, i would not trust that card
Well, toast, or just failing
it's possible it's just messed up the headers or similar, i've seen that before on unsafe shutdowns and the disk is fine, but the #1 priority is to salvage the data before messing with it
ok taking a copy now. THanks
can anyone help with my problem? i don't need to rescue the data but reinstalling the XU4 is a bit of a pain for me cos i don't have an eMMC reader so i'd have to re-image it from a live environment
https://imgur.com/a/AWL3j8t are these error messages expected or are these files in docker corrupted?
that does appear to be not good
can't actually read the error message, can you scroll to the right?
if it's Permission Denied, then that's sort of expected
Yeah I think it was permission denied as I just tried to go to docker folder and it said that and all i see is empty
that is expected, yes. @humble mirage know if it would break everything resetting permissions on those folders to back up?
the docker folder seem to be all empty but the supervisor folder seems to be intact
In the supervisor there is a backup folder with numerous tar files
those are automatic backups. definitely salvage those
the docker folder probably holds the images that can just be downloaded again, but don't quote me on that
so what should I do now that I have backup?
can you plug whatever PC/Pi you were running HA on into a screen and see what happens when you try to boot it?
ok will try and get back to you.
if it's a Pi and it doesn't give a colour splash after a few seconds, then the boot partition is problematic
thanks will get back to you in a little bit
good news kinda, it seems to have booted up fine ..(oddly) and through my browser I got a login screen, but it wont accept my password, dont know if my credentials got corrupted or i simply lost my login info. Any way to reset it?
I can't find that on the bot anymore tink
thanks @humble mirage and @rare ivy I am in, somehow the SD disk decided to work again and I was able to reset my password using external screen and keyboard with CLI line. Thanks guys, you saved me hour and days of headache!! Now on to figuring out what happend to my ISY-994i, the lightening might have blew the board. The lightening strike last night was actually in the local news, it blew apart someone's chimney!!
Thanks again guys
Now is a good time to set up backups
(and they're only backups if they're off the computer you're backing up)
so backons?
backdowns 😛
you mean I shouldn't store all my backups on the device I don't want to fail?
seems inefficient
I know somebody who once set a backup server to back itself up, to itself
That ended poorly, pretty quickly
Tink half the IT world does foolishness like that
Only half?
and wonder "why is the backup of the 100mb files 6.78TB"
Like the OS auditing subsystem which when set to log writes, included the audit log location in that filter, so logged the fact that it logged a write to the audit log
actually had that problem with Avaya PBX's they would backup the backup dir too
what is the proper way of doing a backup, just critical files like the data folder or should I take a duplicate image onto another SD card, (what is the best tool, to do a duplicate image?)
and in the system tools, there is a tool for backup, what files are being back up and what is the method for restoring?
For HAOS you can use #add-ons-archived to push the backup to the cloud, or mount a network file share
For other installs you've got more options
I mount a network file share and backup to that
works pretty fast considering my backup size is like 671mb
i mean, this is better than "why is the backup of the 6.78TB files 100MB?"
that happens far more often than sysadmins want to admit
more often than not it's more like "wait, did we setup a backup yet?
ok but then what is the way to restore if you can explain in general steps for an overall understanding?
"of course we set up the backups! 18 months ago!"
"so that's why there's 17 months of unread emails saying the backup has failed every night?"
"..."
you literally install a fresh os and during first boot it lets you upload a backup to restore from
although if it's a big backup iirc it's best to copy it over after setting up a fresh install? havent tried in a while
should be in the docs he linked
I know a company that had that happen, and found out when they'd lost everything
thanks, that is nice to hear, that it is a simple step
yep
i have haos backing up to my local nextcloud instance using an addon
although now it would be easier to just use an nfs share
Smart I had a question for you and I old manned it
mine was backing up to a USB SSD but i just moved it over to an NFS share on my NAS, which gets backed up to tape
lol
ok, Back to my original problem, I was able to see it on my test station, now I plug it back into its proper location and it boots up, i see it in my router being connected but I can't log in, and when I ping it , it says host is is down!
i think it's actually cheaper to have a ssd/nvme nas than buy tapes these days
shit that reminds me I have a tape drive somewhere
i have a 24-slot tape autoloader that i basically got for free
ya but arent the tapes themselves just going up in price as time goes on?
and even if they arent, isnt it so slow to both read and write that you're basically just doing it for the lulz?
depends if you stick with the generation, then they generally go down, and tape will saturate the SAS interface
so it's just slow reads if you ever need the data?
nope, a restore will also saturate the interface
LTO-4: Generation 4 products reached the market in 2007 and provided a native storage capacity of up to 800 GB and a compressed capacity of up to 1.6 terabytes (TB). The maximum native data transfer rate was 120 MBps, and the maximum compressed transfer rate was 240 MBps.
you using some new shit?
personally, LTO-5
we have LTO-8 and -9 at work
those'll saturate 10Gb ethernet
doing a full restore from tape will basically use all the bandwidth it can get its hands on. what's slow is picking out a single file from that backup
tapes have surprisingly high sequential speeds for reads and writes, better than HDDs, but their random reads/writes are dire
are you sure the router is showing an up-to-date view? are the lights blinking on the network interface?
the thing that is absolutely slow, though, is the iSCSI card i got for my library. i can max out Gb ethernet with the drive easily enough direct through SAS, but the iSCSI card barely manages 10MB/s. i contacted Dell about it and even though i have a supported combination, they don't want to know -_-
trying to diagnose now.. strange.. the problem with lightening, it may have also made the hardware unstable, so checking everything
lol
do you have surge protectors on your electronics? if not, this would be a good time to invest in them
Do have cheap power strips with "surge protection". dont know how effective they are against lightening strikes
depends how old they are. cheap surge protection wears out over time. if the SURGE or PROTECTED light is no longer lit, the protection isn't working. i have several like that. if it's working, then it should protect from lightning. i have some more expensive Belkin units that claim to still work after ~10 years
https://imgur.com/a/0MH1quN. somehow it shows it booting up but not getting ip address
type 'network info' at that prompt
wired or wifi? if wired, are there lights on the port?
Did network info and came back nothing interesting. Listed all the interfaces with not data , or blank
Yes the port lights are on and seems like the router is seeing it
I even plugged in a wireless dongle and came back empty address also.
Booting same hardware setup with a standard RPI boot disk that I know works and see what happens
well nothing interesting to you, but could be helpful to us....
Is a DHCP server on that network?
Sorry. Just a pain to upload photo here. I will capture it the next time.
Yes, fixed and picking up the right address on router. Again it was previously able to pick it up before and I signed in..
Have you tried to turn the router off and on?
I just was able to boot into the same hardware with a different boot sd card .. a raspian os. And was able to ping into it fine,
I don’t trust this sd card or image anymore. Can some guide me how to start fresh and load the backup into it. Can I do this headless? Or do I need keyboard, mouse and monitor and can I load the backup through thumb drive on RPI?
Do you have a downloaded copy of a backup?
You'd have to reimage a new SD card, and then during onboarding you can upload the backup to restore from
Yes have a copy of the data partition files. Don’t like 1 hour ago
It has been like 5 years since I imaged fresh HA. Remind me if I can do this headless or is it through GUI or text based ? I have wire keyboard and monitor but no wired mouse
you won't need a mouse
And do I upload the backup from a usb thumb drive?
im not sure if you can upload it from a thumb drive specifically as I've not had to do it since it was introduced
but I know you can upload a backup
Is there a link to write up about the restore process ?
As you don't have a backup tarball, it might not be as easy. Never did a data partition transplant
you can upload the backup from the machine you are using to onboard HA
Or did you do a backup in the web UI and downloaded that? 🤔
Here is the screenshot of the network info https://imgur.com/a/husbYqc
thanks will be follwing these good instructions
question, so I do have to decrypt the tar files to individual files for upload or do I upload just the one tar file?
What are you refering to exactly "entire backup"? I assumed just the latest one tar file from the backup folder
not sure what you mean by GLHF
If you are using HAOS you can upload that tar file. With other install methods you might have to extract stuff from it.
thank you!
it means good luck, have fun
aka I'm not much help with that method of a restore
yes that's what I mean, the entire backup file, as opposed to having to extract it manually which some users will have to do for their install method (not HAOS)
GLHV. HHAHA. not yet, wiill celebrate when it is up and running!! I am at the upload backup stage, before I choose the file,I noticed the backup tarfiles, are all different sizes some in like 50MB and some are 10KB , my last backup which is like 3 months old is only 10KB, I cant imagine this will be the whole backup, what am I missing here?
wish the instructions were more specific on which files to upload for restoring backup! I can't believe my backup is only 10kb!
my first ever backup was larger than that, and I did it right after I setup
so I'm not confident that's the 'right' backup to load
OK i am gonna try one that is 19MB fingers crossed!
my last backup was 691mb
either your system is huge or obviously I am missing something!!
it seems like the backup files are complete/differential. including system updates. So maybe before os update it does a full copy, but with a fresh image I assume the OS does not need to be updated just the configuration files. This restore does not make much sense to me. I would think restoring configuration should not be that complicatied or big files. But hey I am not a noob!!
yeah you can have full or partial backups
most addons do a partial backup before they update for example
I don't do that, I only do full backups
yeah exactly, and addon backups are different files than system backups, so how do you know if you are restoring correcly by just uploading ONE File!
well I name my backups with the date it was made and _full in the file name
Okay, so the good news is that after a good day and a half to two days of testing, Home Assistant in Proxmox is rock solid. No freezing or crashing issues like I had before. The bad news is that I can't get anything to transfer over to Home. Even adding HA as a bridge in Homekit just hangs and then fails. Still, progress! I think I'm going to start from scratch with HAOS, and hold off on installing things like Scrypted, etc, until I've got just the basic things figured out. 😂
Hi all, for a small HA deployment at the summer house, would a Pi4 with a 32gb USB flash drive be enough?
I guess it's 32gb or more so the size should be fine. Though I am not sure to what bottlenecks will I get myself by using a 10 year old usd stick
It's a HyperX
Any thoughts?
Enough for what? Do you already own the pi?
Yes and the drive
I will just integrate with my heat pump, weather station, a few Shelly switches and a power meter.
Not planning on any computationally intensive tasks
In that case I don't see why not. SD cards or sticks are usually not very reliable (just google raspberry corruption) or fast anyway.
32GB is the recommended minimum so that's fine. We usually recommend a SSD on a powered USB hub for the reasons above.
I'll take this under consideration and maybe sacrifice my 120gb SSD in an enclosure
I think it's an old Samsung 850 or there about
Won't have a powered hub so will have to sacrifice that stability benefit
Pis can't seem to reliable power a SSD on its own so I wouldn't recommend that
Even with a large original power brick?
I don't know why exactly it's not up to the task so I can only speculate. I don't own a pi myself any more and people usually are out the door once they fixed their problem so there's little additional feedback besides that fixed it, if at all.
All I know is that I remember at least 4 people whose various file-system related problems were fixed by using a powered HUB and that the long-term supporters here suggest the same.
So I guess it's back to the 32gb stick for the moment
If it fails I'll build it anew with something more robust
Just make sure you have some kind of off-site/off-main-storage backup: https://www.home-assistant.io/common-tasks/os/#copying-your-backups-to-another-location
I am planning on settings it up with tailscale (which I have for my network) and having it offload through Samba onto my Unraid at the other location.
Thanks for the beauty features, wonder if i can boot the haos into my laptop which does not have ethernet port ? please help
Not really recommended but you can try
- https://github.com/home-assistant/operating-system/blob/dev/Documentation/network.md#enabling-wi-fi
- https://github.com/home-assistant/operating-system/blob/dev/Documentation/network.md#wireless-lan-wpapsk
- https://github.com/home-assistant/operating-system/blob/dev/Documentation/configuration.md#network
I have a question about a Command Line Switch.
I want a switch that if set to on, sends an SSH command and if the command is executed, sets the switch back to off.
As an example:
Switch ON -> ssh test@test 'apt update && apt upgrade -y'
------- Wait for installation ------
All updates installed -> Switch OFF
The switch on command is easy to create but how can I do the switch off?
I want the switch to be set to off automatically after the command is finished.
Howdy howdy - not certain if this is installations or general given it's about core functionality. I just finally got proxmox set up and working and stuck my HA environment on it and I'm not noticing these spitting constantly into the console - can anyone explain what this actually means and what I should look at doing to resolve it?
https://imgur.com/a/c64o7Wi
"[ some kind of string] systemd-journald[115]: Failed to rotate /var/lib/longassstring/system.journal: Read-only file system"
I note that some of my devices seem not to be initialising - I've just got this thing up and running so there might be some faulting services in HA that are causing this
I'm ALSO getting this on trying to do
login and
docker logs hassio_dns
[ERROR] plugin/errors: 2 . NS: dial tcp 1.0.0.1:853: i/o timeout
[INFO] 127.0.0.1: port - port "NS IN . udp 17 false 512" NOERROR - 0 30.00etc
Hope that journal location isn't private xd
No repairs available, by the way.
I might be getting multiple errors or faults due to the same thing, or getting several different ones - I'm new to both proxmox and linux so errors, console messages vs. popups and all of that is still very alien to me. Sorry if this is noob level to you!
I'll add some extra in-system screenshots but of course understand if those aren't related to the above and so need to go elsewhere! Just reply or ping me when you reply by the way and thanks in advance!
Adding; looks like I have another network issue. Stuff can't get online...
Hello, I want to make yearly monitoring of temp and humidity using some zigbee / zha sensors, Is HA good at displaying long charts, or should I use something like grafana ? What database should I use ?
Timeseries DB like Influx or VictoriaMetrics are good choices w/ Grafana... folks in #analytics-archived might have more details, too
It works ! thank you
hey, I am trying to migrate my docker installed HA to another system. I copied the complete folder including the database and docker-compose file. After restarting it on the new system it looks good in the logs and the fronend works. But it doesn't allow me to login
are the user + pwd salted?
Did you remember to copy all the hidden files and folders?
i used rsync -aP so I think this should be fine
the log also say "invalid authentication"
I've cloned my entire config folder to other hosts before and been able to log in just fine
hmm really strange
just logged into the old system to be sure my saved password is correct
ah .storage/auth was not accessible for the user 🙂
just missed the warning during the sync
I am trying KVM installation of home assistant but problem is there are too many network sources to try from and I am getting failed on network manager wait online start
any ideas how to find the correct network bridge (ubuntu)?
oh maybe the problem is using wifi ?
Wired is the right choice for a server
okay, so i went the nuclear option and reimaged my XU4. freshly downloaded 10.4 install onto the eMMC. however, when everything is hooked up, i get
[WARN] Home Assistant CLU not starting! Jump into emergency console...```
it has an IP address via DHCP but no open ports (again).
any idea what could cause this to happen on a blank device?
It was plugged into the network and has Internet access?
You're not doing any filtering, redirecting, or anything to make things more "secure"?
nothing in particular, no. HA has internet access and is on my general-use VLAN
even more interesting, it didn't recognise the USB keyboard i pugged in. had to reboot the XU4
okay, this time i get the HA ASCII art
web UI loaded. okay, panic over
🤞 the backup restores...
i started the restore and the web UI switched off. it's still on its DHCP IP rather than the static one configured in the backup. should there be any backup progress somewhere? i have the CLI open
okay, first restore did not work. i had to upload the .tar through the web UI and then restore it through the CLI. then for good measure, it restored most things (such as my SSL certificate) but did not restore the IPv4 config, which i had to delve into the CLI to fix. oh, and ha network update eth0 did not work for me, it kept giving errors about unrecognised option: device so i had to login and use nmcli
Can anyone help me? I bought my HAOS machine off eBay, pre-configured. It's been working great and I've been accessing it remotely via DuckDNS and port forwarding - I just swapped to a new router/switch and now I can't figure out how to get back into HA.
I had SambaShare, but it seems to not be showing up as available for me to connect to via my desktop network. I can see the computer in my network control panel, see that it has an IP, but when I try to ping that IP or browse that IP it says unreachable.
At this point I'm just trying to figure out how (if?) I can connect remotely to disable the DuckDNS external routing so I can reconfigure. What do I do here? Do I connect a monitor/keyboard/mouse to the HAOS machine? Will that give me anything? Even a CLI? Not really sure.
A keyboard + monitor is a good choice
What am I going to see when I connect? Linux?
A command-line where you can run commands to figure out things like which ip it is set up to use and which method to get it, for example.
And yes. HAOS is based on linux
The console says "a start job is running...* And then eventually says FAILED to start network manager wait online"
Actually now it's on HA CLI waiting for supervisor to start
Try supervisor logs and network info
Network info I think says I don't have a valid IP but I don't really know how to read any of it lol
Make a picture and share it
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.
My docker IP is 172.x.x.x, which I know is invalid but is that right?
That's fine. It's the container network
Are you plugged in via ethernet?
I have Ethernet connected yes
The Ethernet port is showing solid green and flashing orange on the PC and green on the switch
So I have an Ethernet port in but I'm not getting an IP?
It's set up as auto so should get an ip via DHCP. You can try to set it staticly for now
ha network update enp1s0 --ipv4-method static --ipv4-address 192.168.1.230/24 --ipv4-gateway 192.168.1.1
Replace the ips with your network details.
If that gets you access you can configure networking via the UI or leave it as is: https://my.home-assistant.io/redirect/network/
Or run this but this didn't seem to work for you ha network update enp1s0 --ipv4-method auto
I keep getting "context deadline exceeded" when I run commands up update the network
Can you show me a picture of that? Also one of supervisor logs
Actually it worked, I guess my command just wasn't complete. It's showing a static now
Reboot and see if you can access it via the newly configured ip. But if DHCP doesn't work there's probably something wrong.
We can probably tell you more after seeing the supervisor logs.
Something wrong with HA or my network? Cuz everything else got an IP no issue
Don't know. Didn't see the logs and networking is complicated. You can try plugging the ethernet cable to your laptop or something and see if it gets an ip and in what range.
Well I'm staring at my HA dashboard now so, that "worked".
The cable I'm using does assign IP to my desktop, same cable so it's not that
I'd try to set it up to use DHCP again via the link above.
Might just reboot the router after the wife goes to sleep, maybe it's just stuck in a routing table somewhere
There could be a MAC filter or something as well. You can never know 😄
Thanks for your help, at least I can access now while I troubleshoot. WAF restored
My HA is running at 100% CPU load since 2023.7.2
It looks Python3 process is eating my CPU
It happend after one of the recent core updates
I run HA in Proxmox VM on i5 NUC 11th gen. and Python3 in HA shows 400% CPU usage
Heyo all! Just a friendly gentle bump of this once because I posted it in the nini times of the America hours :)
Can you run df -h in the VNC console of the VM? Enter login first
Is for me sir? 👉 👈
Thanks dude, adding that screenshot to the link
Hopefully tha helps, doesn't make a lot of sense to me
I had to try to squeeze it inbetween spammed journal errors
Looks like file system corruption. I just wanted to make sure you were not running out of space or something. Does the datastore in proxmox have enough space available? I'd probably recommend a backup, reinstall and restore. Basically detach the disk, import the ova and then restore from your latest backup
Interesting! Yeah I gave it the default 32, then later added some more to it. Unless doing that caused the corruption (doesn't feel right for a VM host to cause faults with in use disk when adding more but maybe proxmox isn't as clean as vmware?)
Tbh this is a backup of a system running on baremetal which was a backup of a system running in virtualbox... It's probably time to start from scratch
Do you think that'd fix the network connectivity problems in HA as well?
Giving it more space is fine. HAOS will grow the FS. It's just that you can allocate more space than you have to a VM and if it actually uses it you're in trouble
Hmm... I only gave it a max of 96 gb on a 500 gb disk - I wouldn't think it would have run into any problems
guess it's possible though!
Once you got that running I (or someone else who's available) can help you troubleshoot the networking issues
So do you reckon I should just blast it away and set it up from scratch? Think that's my best bet?
Were it windows, i'd probably be thinking so
Also check dmesg -T. Might contain something useful about the IO errors
aa
massive spam
Hold up I snapped a quick screenshot of some of it
See if this helps at all
My recommendation is to power it off, detach the disk and then import the OVA and boot from that. Then restore a backup that was not stored in the VM. No loss of anything that way. In the future you could just restore the whole VM if you have backups set up. Good idea to have both though
Hmm, okay. Let me see what I can do
Thank you for your help!
Yeah; if I can actually get this thing working reliably for once, I'll take and keep full backups properly for DR
Is there an explanation how to clearly use it?
Click the profiler.start link but no
i have generated two files calling the service
but have no idea what to do with both of them
do i need to install SnakeViz and Gprof2dot?
SnakeWiz is enough and easier as it has no additional dependencies
After i have opened .cprof file with snakeviz, how can I find the source of trouble?
Read the Interpreting results docs here: https://jiffyclub.github.io/snakeviz/
Some programming background helps, of course
I have identified following Method in in python3.11 causing trouble
~:0(<method 'poll' of 'select.epoll' objects>)
Do you know what can it be?
That's not helpful but check your PMs
Can't really see much either. It's probably one of your integrations
Hi, I have bought myself a Rpi 4 B and I have written the Home assistant os onto an Kingston SSD and I have written the USB bootloader onto the micro sd card.
With everything plugged in (SSD over usb, micro sd card and Ethernet) the green ACT light is still flashing even after waiting for 30+ minutes.
Does anyone know how to fix this issue?
Can you plug in a keyboard and monitor to see what's going on?
The Pi4 should be able to boot directly from the SSD
Possibly... 
I have tried to install the home assistant os already before that but with the Ethernet cable not plugged in so I pulled the power cable, connected Ethernet and connected the power again so maybe that changed something to go wrong
But I rewrote the os onto the ssd and the bootloader to the micro SD before Plugin in the power again to be sure no files are corrupted
So if I plug in the ssd without Plugging in an SD it just bootes from the ssd?
To configure your Raspberry PI 4 to boot from a USB drive, you just need to enable this feature inside the boot code, stored in EEPROM. And here’s the best part: Most Raspberry PI 4 boards enable this feature by default! Only a small batch of early Raspberry PI 4 boards shipped with this feature disabled.
How can I acess the boot code / EEPROM?
Only a small batch of early Raspberry PI 4 boards shipped with this feature disabled
You almost certainly don't have to
If you do, Google is your friend
If the SD is still good then just move it over and the Pi will boot up fine
They're great educational toys 😛
just clone your SD card
to another one
and have a spare pi
that's what I did when I ran on a pi
I have since seen the light
if you use the same sort of SD cards cloning one to the next is easy
and then you can always restore a backup to get "current"
that's not quite the same as what I meant
I meant read the SD card to an image and flash that image to the card
copying the contents might not get all hidden folders etc
Win32DiskImager is good for windows
But that link has it all
ddis 👑
every time i tried linux fs on windows it ended catastrophically. I don't touch it anymore
Restarting the USG or UDM sometimes solves DHCP issues.
yes the UDM likes to be reboot
I don't have any of these issues you do, no UDM, no rpi4 and no wife LOL
Now I only plugged in the ssd with home assistant os on it and it still blinks continuously
Although the blinking seems a little slower than before
hello, has anyone encounter a error when onboarding. i used f12 to check for any errors on my browser and im getting this error "homeassistant.local:8123/manifest.json 404 (Not Found)"
any help would be great appreciated
I am at a lost. I am not the smartest, and im sure its something simple, but I can not get HA to install and run on anything. I am currently trying to get it installed running on a windows machine in VMware. Everything seems to load fine, I get to the PREPARING HOME ASSISTANT, and thats where it always fails.
5 hours?
Started the last install at 3pm, almost 6pm my time
its almost 6 pm now. So almost 3 hours on this install.
I've seen it take that long
I just grabbed a PI4 out of the closet that Im not using. Going to attempt to install it on that, and I will let it run overnight and see if I have better luck.
Atleast that will eliminate if its a problem with my VMware configuration.
Just getting lots of Server Errors... and timeouts
How's your internet, and your DNS? Are you wired or wireless? Problems are at least 50% DNS hahah
decent internet, dns is flaky lol, wired
Why is dns flakey? Are you using local dsn resolver? or your ISP? Or public (google, cloudflare)
i'm sure it's been mentioned, but with the latest build of HA, I can't use Chrome to set it up. no biggie, edge works fine.
public, I didnt have my damn network configured right in the VM. Its working great now
Update: Restore finished after 1h21m!! Sorry for the interruption.
Hi all, I'm attempting to move from a Pi4 to an x86 machine, and (I think I) am having trouble with the restore process.
I imaged the drive on the x86 box, and was able to get to the onboarding page. From there I uploaded the backup, but apparently there is no way to check the status of the restore. I ended up powering down the system, and trying again later. Most recently, I created a throwaway account and did the restore from the web UI, but I'm getting the same results (can't access the web interface, but the observer says everything is ok).
I've checked the CLI, and, according to "supervisor stats", the system is apparently using 100% of one of the cores on the machine, so I guess it's still restoring, but we're at 53minutes now, which seems excessive for a 250mb restore on an i5-6500T (which, while old, should be quite a bit faster than the Pi 4 I'm upgrading from).
Something else of note, I haven't plugged in the various USB devices that I use (SDR, ZigBee, and Z-wave radios) because they're still on the Pi4, and the Pi4 is still on the network.
Is there any way to see if it's actually restoring? Any suggestions? Should I expect the system to take several hours to restore?
My raspberry pi 4 doesn't boot from USB. I tried installing home assistant os onto the sd card and it worked fine but when removing the sd card and only plugging in the ssd with home assistant os on it the green act light just flashes infinitely
Can anyone help me?
Ignore. Normal.
Well this article doesn't really help me since I already did all of that except for connecting a screen in step 6 of flashing USB boot. But I always waited until the act led started flashing so that should be ok
You didn't say what you'd done, or not done, so .... how are we to know?
.
All the way back there, because we do scroll back and read everything you've ever posted 
Well you answered to that question so I thought you knew
Anyway maybe the EEPROM is corrupted?
Idk I read that in a forum post so could that be the cause?
.
I noticed now that there are two places to set the language. Under profile and under Settings > System > General. But only when i change the language via the profile setting the GUI changes to the set the language. Is the language setting under Settings > System > General obsolet?
Default language used by Home Assistant. This may, for example, influence the language used by voice assistants
The UI defaults to using your browser language if there's no language set in the profile
Hmm confusing. My language ind the profile is german and under Settings > System > General set to english, but when i open Voice Assistant the dialog is german.
Right, because the dialog is in the UI
Ist that possible as a language must be choosen during onboarding?
However, if you just call a TTS service, it'll default to English
Anything without a reason to be otherwise will default to English, because you told HA that the default is English
Ok, understood, thx
Hello, I've followed the Raspberry Pi installation guide (I have an RP4), and made it to the onboarding screen. However, I went to go restore from a backup (backup is a partial backup from 2022.6 I believe), and it showed that it was restoring, but port 8123 is no longer responding. I did a port scan and found port is open 4357, which reports via a browser:
Supervisor: Connected
Supported: Supported
Healthy: Healthy
Anyone have any idea how I can recover from this state? Port 8123 reports it's closed.
Might take a while. Try if supervisor logs in the HA CLI has anything interesting/recent
How can I access the CLI? I'm installing it via the HAOS method, not docker. I don' t know how I can access the Pi to retrieve more logs. SSH is closed as far as I can tell.
Also it's been 12 hours 🙂
Connect a keyboard and monitor
I don't have access to a mini-HDMI. I presume there is no other option at the moment based on those facts?
There's a "forbidden" way to enable host SSH: https://developers.home-assistant.io/docs/operating-system/debugging
Can't think of anything better right now
@raven forge Ok, thank you for the ideas
FYI, I alwyas add SSH & samba share to my machines so that I have 2 ways of looking at (some of the) logs.
If HA isn't running but the rest of the containers are, you'd have access via SSH and Samba share
Good idea. After this struggle, I will do that for the future. Thanks.
You could re-flash, install the addon and then try the restore again
Thanks folks. Appreciate the ideas!
Is there an official minimum-spec for HomeAssistant OS? Aka could I run it comfortably on a Wyse 3040 thinclient?
It really depends what you plan on doing with it. It's hardly more powerful than a Pi 4, so it'll likely run with no issue, but if you want to do any video processing, etc, definitely not comfortable
milliseconds 🙃
well then I guess it's not so bad, as I usually measure lag in seconds
Gotcha. Just curious. I'm running it in a VM now, but I'm thinking of trying to lower power consumption in my apartment but /shrug/
Depending on your budget... the N100 mini PCs are quite capable w/ low power usage
also, modern
my SFF PC draws about 75W and it's much beefier than a rpi or wyze
I used to be that guy that swore by rpi's, now I only swear at them
Hi Folks, sorry in advance if this isn't the right place to ask for help with an issue but I'm currently stuck unable to upgrade my HAOS to 10.x and not sure how to fix this. I've raised an issue on github but its gone a bit stale and I need some help https://github.com/home-assistant/operating-system/issues/2522
There isn't a clear migration path from supervised to container, is there?
Please use a code share site to share code or logs, for example:
- https://dpaste.org/ (select YAML for the language, and consider picking a longer expiry)
- http://pastie.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.
You can copy the configuration over manually if that's something
You mean the whole hassio directory?
Usually the config folder but I stand to be corrected
You'd also have to spin up your own containers for the additional things
Was that for me?
Yes
That I know, it's the HA instance itself that I'm worried about.
Well you'd lose the supervisor going that way
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.
Yes, I'm kind of done babysitting it.
I have a supervised installed from years ago and it's not happy that I'm maintaining a lot of other parts of Debian separately outside of HA.
Is it really that much? Seems high
Which sff? My thinkcentre are about 10W
Hi, i noticed a few times already that my HA reverted to an older version when restarting HA.
Automations, dashboards, settings disappear or reappear.
Am i doing something wrong?
I just restarted again.. and its back to the new version.
What's your install method?
I am runnning a dedicated laptop, not sure what the versions were called tbh, core and supervisior would be my guess
if i understand the question correctly
The topic of this channel is:
Discussion about how to install Home Assistant (https://www.home-assistant.io/installation), and issues with the install itself. Make sure to mention which type of installation you have. If you're not sure of your installation method, check under Info in the System Health Card (https://my.home-assistant.io/redirect/info/)
Use the link there to find out
Then it would be Home assistant OS
@eager geyser 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).
Then I have no idea why it's misbehaving that way, maybe somebody else will
A host reboot can't hurt
That does sound a bit like the behaviour I'm getting with HAOS. I have been trying for ages to upgrade from 9.5 to 10.x but it never works and I usually end up rebooting and finding I'm back on 9.4 when I was on 9.5
Might want to take a look at the supervisor logs
i noticed the same this morning, with the update of 2023.7.1 to 2023.7.3, but tried again and it that worked.
I will have to keep a close attention to the updates in combo with restart.
I will have alook at the supervisor logs as well, thank you for the input!
Lol, ok.. just noticed.. my update worked fine.
But my system got reverted again.
Restarting again..
After the update everything is up to date and i am back to the correct version of HA
My supervisor logs dont show any issues, only info messages
Hi everyone, I'm hoping someone can help me. My HA installation broke yesterday and I can no longer access the HA front-end, but automations and Alexa control still seem to work.
I had a working Home Assistant installation in a Docker container (using Docker Compose), accessible from outside my home network using Nginx Reverse Proxy and a personal domain I own. Two days ago I was attempting to add another image to the docker-compose.yml file, and noticed after starting up the containers that I can no longer access the HA frontend at all. If I try to navigate to the server IP and port that HA runs on, my browser says it can't connect to the server. If I try to connect using the hostname I defined in nginx reverse proxy, I get taken to a page with the Home Assistant logo telling me it can't connect and will attempt again in 60 seconds. If I wait and let it try again, I get taken to a generic "502 bad gateway" page like when I try to connect via IP and port. I then stopped my containers, removed the docker image I added, and then started up again, but the issue persists.
I run several other docker containers which also use nginx reverse proxy and they're all working fine. I've taken these steps to try to resolve the issue:
- Confirmed Nginx Reverse Proxy settings were correct for the HA hostname
- Renewed SSL certificate (LetsEncrypt) through NginxProxyManager
- Restarted HA container
- Rebuilt HA container
- Restarted host computer
- Started over with just a docker-compose.yml file and let HA rebuild itself
- The same step, but running
docker system prunefirst to make sure all unused/stale containers/images were removed
I found several threads on the HA forum with things to try, but none of them worked for me. One that did, however, is that I can open a Bash contianer into the container HA is running in, and ping 127.0.0.1:8123 and get a response. So ultimately, it seems like what's going on is that the HA front-end isn't reachable from outside the container, but HA itself is still working.
Could anyone help me diagnose this further? I've hit a wall and I'm pretty desperate at this point.
docker-compose.yml
version: '3'
services:
homeassistant:
container_name: homeassistant
image: "ghcr.io/home-assistant/home-assistant"
volumes:
- ./config:/config
- /etc/localtime:/etc/localtime:ro
restart: always
privileged: true
network_mode: host
I did, one sec
Did you try connecting directly to HA on the IP and port?
Sorry, I did a lot of research the past 24 hours 🙂
I did, my browser says it can't connect
@frosty hawk 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).
The log file should hopefully help
Please use a code share site to share code or logs, for example:
- https://dpaste.org/ (select YAML for the language, and consider picking a longer expiry)
- http://pastie.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.
Sorry! I'm a bit sleep deprived
...that was muscle memory
so sorry
lol
Well, don't do it again please
won't happen again
So strangely, the home-assistant.log file (and home-assistant.log.1 file) are empty, but docker compose logs shows this:
homeassistant | s6-rc: info: service s6rc-oneshot-runner: starting
homeassistant | s6-rc: info: service s6rc-oneshot-runner successfully started
homeassistant | s6-rc: info: service fix-attrs: starting
homeassistant | s6-rc: info: service fix-attrs successfully started
homeassistant | s6-rc: info: service legacy-cont-init: starting
homeassistant | s6-rc: info: service legacy-cont-init successfully started
homeassistant | s6-rc: info: service legacy-services: starting
homeassistant | services-up: info: copying legacy longrun home-assistant (no readiness notification)
homeassistant | s6-rc: info: service legacy-services successfully started
If you're having problems with your updates to your configuration:
- Check the troubleshooting steps
- Check your log file - remembering you may need to set logger to
infoordebug - Explain what the problem you're having is - sharing configuration, errors, and logs
Set logger to info and restart HA
There we go, lots more info available now, but nothing jumped out at me as being problematic. Let me attach the log file
No mention of HA being started, or the frontend starting
Oh, no, there we go
2023-08-01 08:40:51.719 INFO (MainThread) [homeassistant.components.http] Now listening on port 8123
complains about "SmartTV 4K FFM" tho
I'd also expect
2023-02-15 14:53:17.264 INFO (MainThread) [homeassistant.core] Starting Home Assistant
Oh, that's there
2023-08-01 08:41:22.601 INFO (MainThread) [homeassistant.core] Starting Home Assistant
You and me both, lol
sudo ufw status shows that it's inactive
do you have default_config: ?
Yeah, right now my configuration.yaml is fresh, HA rebuilt it on startup
....urgh, sorry
I'm really, really trying not to do that
but yeah, other than adding the logger: info portion, configuration.yaml is fresh, and has default_config and all the imports
Tinkerer, does the fact I can bash into the container and successfully ping 127.0.0.1 port 8123 jump out at you?
ping?
The logs do say the server is listening on port 8123, so I guess it shouldn't
You can't ping a port
Yeah, not ping, what's the other command...
I mean, you ran it
curl
If you don't know what you're doing... how are we going to know 😛
And on the host, can you curl that?
you can't ping a port but you can telnet to one 😉
Nope, that fails
Waits for you to telnet to your DHCP server 😛
also curl is good
Then... Ubuntu is up to the usual Ubuntu games
Sounds like you've dealt with this before
it seems like HA isn't accessible from outside the container all of a sudden
I've seen many people bit by Ubuntu making some terrible choices
Perfectly fine desktop OS, but as a server ... 
It's what I've run my server on for ...2, almost 3 years now? But I guess nothing's perfect
The strange thing is that installing a new docker image (wyze-docker-bridge) seems to be what prompted HA breaking
I didn't see any issues over there related to what I'm experiencing though
Have you tried excluding that container again to be certain about that? Also, for fun, have you tried rebooting the host all together just to see if there's something silly going on that might just be resolvable by that
well telnet is for testing TCP ports not UDP, and DHCP expects a broadcast anyway
NSX, yes to both
and HA is available once the bridge container is stopped?
no, it's not - even after stopping the bridge container, uninstalling it, uninstalling both and trying to start fresh...the problem persists through all of that
So 🤷♂️ may or may not be related to that then
Yeah, Tinkerer pointed out that the logs show that HA is up and listening on port 8123...going by the logs alone it looks like HA should be all good, but I just can't reach it. Even if I try to access it with the machine's IP and port on my local network
sudo netstat -tulpn | grep 8123 on the host, anything there?
nope, nothing
But it returns back results if I exec that in the docker container
The HA guide recommends specifying network_mode: host, so shouldn't the port from the container be inherently accessible to the host?
Should be as long as you've got host networking
Hi all, today I can't access my HA server via Nabu Casa. It was working yesterday and my home internet still works. I know my HA is working cause I used it this morning before I left for work. I coincidentally received a receipt for the next 1yr of Nabu casa subscription. Is it possible that on subscription renewal days, Nabu casa stops service for some period of time?
I'd check in w/ #cloud-archived for Nabu issues
Thanks!
lol
Pigeons are delayed more today
the dammit was at NSX beating me again, not you Kevin
🤪
my pigeons are about to unionize I think
Perhaps just to confirm check docker inspect homeassistant -f "{{json .NetworkSettings.Networks }}"
Assuming homeassistant is your container name
Anything in particular I should look out for there?
It'd start with {"host": { ...
Yeah, I got a result back from that
some fields are blank, like the IP, gateway, and Mac fields
{"host":{"IPAMConfig":null,"Links":null,"Aliases":null,"NetworkID":"3508b0395c2422f303ba1a36c354e09fa2a10204f39f2e56b603c13f5632ef7a","EndpointID":"540b3dcecea0add6d7ca0619cb224bd19b89aa31a939af627a33ea8d51a748d3","Gateway":"","IPAddress":"","IPPrefixLen":0,"IPv6Gateway":"","GlobalIPv6Address":"","GlobalIPv6PrefixLen":0,"MacAddress":"","DriverOpts":null}}
You think it failed to generate a network for the container?
Those are empty for me too though, running HAOS
Actually I think that's normal, it's running in network: host mode
so the host is the network...I think
tbh HA is the first time I'm using that, my other docker containers all run bridged with ports explicitly declared
What did you see when you ran the netstat command in the container
tcp 0 0 0.0.0.0:8123 0.0.0.0:* LISTEN 66/python3
tcp6 0 0 :::8123 :::* LISTEN 66/python3
Yeah, I mean everything inside the container looks good...
Just curious, does the container have internet access
I'll try pinging from inside it
Yep, it has internet access
it can ping google.com
The only other HA-related troubleshooting step I can think of is...is there anything inherent to HA's startup process that "registers" it to the host system in any way? For example, I see in the logs there it says "listening on port 8123".
I'm just trying to think of anything else we may have missed...otherwise this seems to be more of an issue with ubuntu or docker rather than HA, because inside the container, it seems like everything is running correctly
This is entirely an issue somewhere between Ubuntu and your Docker stack
Agreed, I don't think HA is the issue here. Thanks for all your help everyone 🙏
Hey folks, I've made some progress in researching a fix. It seems like network_mode: host is the issue here. If I instead declare a port assignment for - 8123:8123 in my docker.compose-yml file, I'm able to connect to the HA front-end again.
Anybody happen to run into this sort of issue before? If not, no worries, I'm still researching on my own. But I'm hoping someone might have run into this before.
I'm not using Docker Desktop and I'm on an Ubuntu headless server (not Windows or Mac, where network_mode: host reportedly doesn't work)
How did you install docker? I'm asking because I've heard of snaps causing strange issues
Should I keep my Home assistent on my rpi 4 8gb with 250gb ssd? Or choose for a intel nuc?
If so what intel nuc would you recommend? It must be future proof as well
NUC > Pi
#hardware-archived has a lot of recommendations - but mostly it comes down to your budget and what you plan on running on it
The requirements are very different if you want to start running things like Frigate and DoubleTake
That'd be #integrations-archived
thank you, I'll move it there
I wonder if Duckdns is down
Hi, i yesterday talked about my system reverting to a previous state.
Noticed the same issue before going to bed.. checked if my HA itself was up to date, all fine.
Restarted "the system running Home Assistant and all addons".
Ah, everything is back to normal.
This morning i checked and the dashboard, automations, newly added integrations were gone / reverted again.
Any idea on what i can do / check please?
Running HA OS on a dedicated laptop.
I just restarted "the system running Home Assistant and all addons" again.
Everything is back.
This happends rly often 😦
I've read that HAOS will revert to a previous version when it fails to boot X times but that shouldn't revert dashboards.
How much free space do you have? A look into the supervisor and maybe even host logs sound like a good idea
i dont see anything that i would say is an issue.
i do see in the host log that HA started at 03:43 this night, so i assume thats when the restart happend.
In supervisor i do see something about space and about autofix:
https://imgur.com/a/aHnWwkC
Disk is only 8% used
Memory 16%
CPU 4%
Last boot.. is last month weirdly enough
Hello all,
I have tried searching for an answer for this elsewhere but I imagine I'm just using the wrong search terms etc. I feel like I've missed a notification or announcement or similar that explains my issue, but does anyone know why the Docker Hub homeassistant/amd64-hassio-supervisor is no longer being updated (nothing since June)? Have I missed a move to a new source etc? 🤔
Thanks in advance for any answers!
What's your install method - HAOS or Supervised?
Is Hass Agent only 64-bit version, since I can't find 32-bit version to install on an old pc.
They have their own Discord, maybe ask them there?
Is there a way to install Home Assistant on my pi that already has other things running on it?
Using docker you can install and run the HA container. It won't have the supervisor/add-on ecosystem but you replace that with your own docker ecosystem
a wahh
indubitably
what's the "supervisor/add-on ecosystem"
other software outside of Home Assistant itself. things like node-red, mqtt brokers, etc
might include things your pi is already running, might not
pihole used to be an add-on but it was deprecated. adguard is still available though
That's a mighty pi!
(and it's not an official pi either, it's an Orange Pi 5)
plex and jellyfin are available as add-ons, but if you have it already...
ah, i see, it's programs that i could just run outside of home assistant
oh if it's an orange pi you would want to run the docker installation method anyway, regardless of whether you had other things on it or not
yeah ik, it doesn't like standard OSes
yeah there is not an HA OS image for orange pi
I'm in the process of swapping the 500GB ssd for my 8TB hard drive, but that requires me to copy all the 4TB of data on it over to other drives to change the file system to ext4
then copy it back
and i'm struggling to find 4TB of free space
that isn't on my big storage drive
Docker is a great system to get familiar with anyway, have you used it before?
kinda, on caprover
cool
that runs a few things too, like my website, and my MC server's control panel's rev-proxy
yep, most everything on my home server runs in a docker container
yah, caprover is running on oracle cloud :)
well i'll be able to figure this part out, i just wanted to know which installation method i should use :)
and if it had to be docker or not
So i follow the "Install Home Assistant Container" method?
yep
cooool
docker is like a jukebox which plays all your software "CDs". Home Assistant would be one of those. You can add and configure others and self host tons of stuff.
yea, i use it on caprover and it's so cooooool
and ofc i'm currently going on a sidequest of getting ssh keys setup on my server cuz i should probably (have done that a while ago) do that
oh dammit i lost the public key i had lmaoo
nvm found it
the .8 update added a warning about my mqtt device namings
what is the reason behind it
You might want to share that warning 😉
Recognized entities with a name starting with the device name
i am a bit confused because it lists entities likesensor.hm_600_yielddaywhich are looking similar to the default ones on my non mqtt devices
oh i suppose the device name is supposed to be hm_600 and the entity of the sensor would be yieldday?
Is it a Zigbee2MQTT device?
no its opendtu
notify opendtu of the breaking change for MQTT HA
unless you create the yaml entries yourself
if you do, follow this -> https://community.home-assistant.io/t/psa-mqtt-name-changes-in-2023-8/598099/14
ah i checked their issue page, someone got to it faster
well guess i observe and update when there is a fix for it
hope i dont loose historic data that would be sad
thanks for the details!
Hello guys. Need help in getting Insteon 2413U PLM going. I used to have a isy994i but it got fried from lightening. So I decided to go with a new 2413. I got the integration into homeasssistant but it is not seeing any of my devices. How do I get the Insteon integration to see the existing Insteon devices ( I have like 30), do I need to manually go around linking it to the plm ?
Hi y'all, trying to install using vbox but I'm just getting a black screen for the VM - i followed all the instructions correctly & double checked them. On Ubuntu
Anyone have an idea as how to fix?
I might use a different vm software
you're trying to load up HAOS as a vm on virtual box with ubuntu correct?
Yes
did you enable efi?
Yep
what are your boot options ?
It's booting from the hard disk (the vdi file)
in the virtual machine config, do you maybe have another device set to boot before the hard disk?
and double checking you did set it to Linux 2.6 / 3.x / 4.x (64-bit) ?
I unchecked the other devices and moved hard disk to the top of the boot order
also did you make sure that your BIOS supports virtualization? and if you did, is it turned on?
Oh idk
Hello everyone, so I need some guidance on getting the HA companion app (ios) working. When I try to connect to my server I get this.
"NSURLErrorDomain -1099"
#ios_and_mac-archived can help
Anyone else encountering an internal server error (500) when their install tries to pull the generic x86_64 docker image?
that's indicative of DNS issues
23-08-02 19:54:31 ERROR (SyncWorker_3) [supervisor.docker.interface] Can't install ghcr.io/home-assistant/generic-x86-64-homeassistant:2023.8.0: 500 Server Error for http+docker://localhost/v1.42/images/ghcr.io/home-assistant/generic-x86-64-homeassistant:2023.8.0/json: Internal Server Error ("readlink /mnt/data/docker/overlay2/l: invalid argument")
23-08-02 19:54:31 WARNING (MainThread) [supervisor.homeassistant.core] Error on Home Assistant installation. Retry in 30sec
Should my dns be 192.168.1.1?
if you have a LAN dhcp server/proxy at that address.
check with nslookup
that probably comes from your dhcp server
otherwise just point it at google/cloudflare, at least for the install
Still having the issue after adding 8.8.8.8 :(
I just realized the update didn’t stick, why is that?
Didn't stick?
if it failed to download, it won't apply
Are you out of disk space?
Shouldn’t be, I have a terabyte
Could it be that I’m running out of space because of weird partitioning?
I burned the image directly to my hard drive
Oh hey, that fixed it!
Trying to install generic x86_64 on a laptop, balenaEtcher keeps erroring.
Weird, I didn’t use balena etcher, I extracted the xz with 7zip and then burned the img to my hard drive using Rufus
"Something went wrong. If it is a compressed image, please check that the archive is not corrupted"
Realized that I had a 10.3 image downloaded, that appears to be working. Something is broken with the generic 10.4 image.
Another question... Trying to boot a laptop off of an external HDD flashed with the HAOS generic image. It looks like it tries at startup and then gives up and boots windows. What is going on??
Also, balenaEtcher still doesn't work with that link above.
I tried downloading the file at that location and was able to flash with that. But that doesn't work.
Is it normal that the image creates like 6 different drives?
Is it not possible to use some USB to SATA adapters to boot HAOS?
Yes, boot order is correct. It tries the external drive, but gives up after a few seconds.
Dang I mistook the wrong option to disable secure boot... Such a fool am I.
it be like that
Hi, I am trying to add Bluetooth integration.
It automatically discovered the integration once. Then disappeared. Since then have tried multiple times but it just doesn’t show up. Tried in built Bluetooth and Bluetooth dongle with pass through.
My setup is
Windows 11 NUC
Virtualbox
HA OS
Bluetooth dongle with 2m cable extension
USB pass through in virtual box
Can someone please help me
#bluetooth-archived can probably help better
I can't seem to install updates. I had this recently and got past it by deleting dangling images and stopped containers but that's not happening this time :/
Here's a Supervisor log which starts with a giant error that I wonder if it might hold a clue:
http://pastie.org/p/4qxtRuRdd336TarusE4jBr
what return curl -v https://github.com/esphome/home-assistant-addon/
?
do you use HAOS?
Yes, HAOS. Just ran that curl and it's showing me the full markup for that web page
Did you run the curl on the HA server, or on your own device
Since the logs seem to indicate your HA cannot connect to github securely
I ran it from the terminal add on.
This is the thing, it's like every few requests from github fail
maybe curl ise other SSL settings that python?
Even pulling the image manually seems to keep needing to "retry" and eventually it just fails with "unexpected eof"
https://imgur.com/a/9We0Ogw
I tried running update-ca-certificates on host, homeassistant and on supervisor to see if that would help but no luck 😦
Oh wow, that latest pull actually worked... not sure if that helps me though.
I triggered the update now and it seems to be installing from that successful pull, hopefully that's the trouble over then. But I don't understand why this keeps happening to me recently 😦
Attempting the OS update now, hopefully it goes smoother.
Looks like the OS update has worked, it's restarting at least.
Pretty sure that docker pull line succeeded only because I'd asked for help on here 😆
Sure taking it's sweet time to come back online though :/
Typical, came back to life as I posted that.
Whelp, all working now. Hopefully it stays that way :/
Hey guys, I'm trying to update to 10.4 from 10.0 or 10.1 but each time I try to update it takes a while, the rapsberry pi reboots and the server is in version 10.0 or 10.1 (sometimes one sometimes the other). I can't figure out how to get the logs of the update to see what's wrong, do you know how I could find what's happening?
to update HAOS*
so umm it worked
One workaround is taking a full backup, installing 10.4, then restoring from backup. It’s probably easier than trying to diagnose why it’s failing
whats the best way to get home assistant running away from my home network
without a vpn
Cloud flare
Home Assistant Cloud 😄
I tried that but I keep getting 400 bad request
Nginx reverse proxy using something like nginxproxymanager or traefik, or cloudflare tunnel where they mitm your traffic
Dynamic dns?
tf is dynamic dns
I just used cloudflare tunnel
but when I went to the link it didnt work
Did you check HA's log file for warnings or errors?
Because you have to configure http correctly
Nice man, good to hear
Okay thanks, I'll probably end up doing that
RPI4 8GB USB Boot failing 3 green flashes
I have a crazy question. I’m pretty new but is there a way I could run HAOS on a synology NAS or is it strictly the Docker version?
Some of them can run VMs
The one I have can. Right now I’m running HA on Docker which I’m missing the add-ons ability. I was thinking if I could run it on the VM, would I get the add-ons back?
I had HA running on a Raspberry PI but I rather run what I can on my NAS instead of having a bunch of things plugged in. The add-ons made everything so easy. I just don’t have the patience with the Docker version.
Worth a try. What can you lose?
This is true lol. I didn’t know if there was any reason I couldn’t or shouldn’t.
After running the docker image I am getting the following error:-
docker: Error response from daemon: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: exec: "/init": permission denied: unknown.
While I am easily able to build the image. But while trying to run it, I am getting this error.
What could be the solution for it?
Yes I am running the commands as root
Anyone know if I can run my raspberry pi on a kingston portable ssd ?
connects through usb c
Maybe some ideas on this here?
Maybe a dying microSD? 
maybe PSU is not putting out what it says it will
I am moving my HAOS VM install to a baremetal HAOS install. Anything that I need to be aware of besides just back up old and restore on new?
nope
if it's a really big backup file you might be better off setting up new and copying over the file via samba etc then restoring from backup
i havent had to do that personally though
https://www.derekseaman.com/2023/04/home-assistant-pt-3-restoring-your-configuration.html whatever he says here is probably still relevant
I have a fixed IP for my VM, probably best to release that and use the same IP on the new box right?
and update to most current?
How long should it take to restore? My backup file was only 200MB
Restoring backups takes forever at moment (edit: for some people). Some users report about 3 hours on a i3 NUC or approx. 6 hours on a Pi4 based system. Some others seem to be more lucky.
Just let it do its magic
There is an open supervisor issue since May, which might indicate in that direction
