#installation-archived
74689 messages · Page 78 of 75
Afternoon all, got an issue recovering from a back up all entities are working but all add-ons are failing to install with an Unhandled exception.
22-03-10 14:48:10 WARNING (MainThread) [supervisor.backups.backup] Can't restore Add-on core_ssh:
22-03-10 14:48:10 INFO (MainThread) [supervisor.addons.addon] Restore config for addon a0d7b954_wireguard
22-03-10 14:48:10 ERROR (MainThread) [supervisor.jobs] Unhandled exception: 'a0d7b954_wireguard'
I just hopped in this discord because I have a similar issue.
Got a new SD card, flashed it, booted into the HA landingpage, and instead of setting up a new user, I selected restore from a backup.
However, nothing really ever happens.
And when I try to access the page it is no longer available
Glad I'm not the only one.
Are you using HassOS? I am.
Correct
Do you run nabu casa or use a different method to remote in when outside your network?
i have it all privately routed thru my firewall and have a few registered domain names
Are you using duckdns?
no
i wonder if i clone my SD card, if there is a way to expand HassOS after the fact.
flash my current image that is 8 GB onto a 64 GB card, then have HassOS detect that and expand
I believe it should happen automatically
cool. i'll find out shortly. haha
Solution: Installed the add-ons then run the restore.
This worked. Now shows the SD card free space, and everything appears to be working. Going to doing some testing to make sure all my automations and integrations work, but looks good at first glance.
I've just tried updating HA from 2022.3.2 to 2022.3.3 and got WARNING (MainThread) [supervisor.jobs] 'HomeAssistantCore.update' blocked from execution, system is not healthy this error

Hoping for some help with initial setup. I'm using a VM on Unraid (KVM), but get the same results with a physical RPi install. Everything seems to be working, can resolve both hostname and IP address, but the onboarding page will not load. (The observer page loads fine.). Have access to the HA CLI, but no idea what to do there. Anyone have suggestions?
Had to re-install Home Assistant OS on a rpi3 with 1 GB memory after a SD Card Failure. Since reinstalling and setting up all of my integrations, my Pi has been crashing due to memory processes being killed. Been rolling back to prior backups that are more stable, but cannot determine what's causing the Pi to crash. The logs aren't telling me much, and only see the errors in CLI. What's the best way to find the root cause?
@blazing sandal I'd guess the 1GB is at the heart of it
Do the prior backups have less going on?
That's what I was thinking too, but prior to the SD card failure, similar integrations were setup and was rock solid. The prior backups have less going on -- mostly Unifi Protect cams
any have hasio OS in proxmox ? after retart server console stop there and nothing happend
https://i.imgur.com/NeU12a0.png
after 5 min vm restarted i just see now.
i dont have backup 😦
never saw this -> https://i.imgur.com/GKEdZqz.png
i think the problem no auto boot.
just need select boot
😮
never happends
hey - a question for supervised docker installation - is it still buggy? I have retried it now and supervisor cannot start for more than 30 min, first due to node-red, but core man it is all error...
2022-03-10 22:05:06 ERROR (MainThread) [homeassistant.setup] Unable to set up dependencies of cloud. Setup failed for dependencies: http, webhook
2022-03-10 22:05:06 ERROR (MainThread) [homeassistant.setup] Setup failed for cloud: (DependencyError(...), 'Could not setup dependencies: http, webhook')
2022-03-10 22:05:06 ERROR (MainThread) [homeassistant.setup] Unable to set up dependencies of analytics. Setup failed for dependencies: api, websocket_api
2022-03-10 22:05:06 ERROR (MainThread) [homeassistant.setup] Setup failed for analytics: (DependencyError(...), 'Could not setup dependencies: api, websocket_api')
2022-03-10 22:05:06 ERROR (MainThread) [homeassistant.setup] Unable to set up dependencies of onboarding. Setup failed for dependencies: analytics, auth, http, person
2022-03-10 22:05:06 ERROR (MainThread) [homeassistant.setup] Setup failed for onboarding: (DependencyError(...), 'Could not setup dependencies: analytics, auth, http, person')
2022-03-10 22:05:06 ERROR (MainThread) [homeassistant.setup] Unable to set up dependencies of frontend. Setup failed for dependencies: api, auth, config, diagnostics, http, onboarding, search, websocket_api
2022-03-10 22:05:06 ERROR (MainThread) [homeassistant.setup] Setup failed for frontend: (DependencyError(...), 'Could not setup dependencies: api, auth, config, diagnostics, http, onboarding, search, websocket_api')
2022-03-10 22:05:06 ERROR (MainThread) [homeassistant.setup] Unable to set up dependencies of safe_mode. Setup failed for dependencies: frontend, cloud
2022-03-10 22:05:06 ERROR (MainThread) [homeassistant.setup] Setup failed for safe_mode: (DependencyError(...), 'Could not setup dependencies: frontend, cloud')
@empty turtle posted a code wall, it is moved here --> https://hastebin.com/dezukejexo
@empty turtle Rule #6: Please do not post codewalls (text longer than 15 lines) - use sites such as https://www.codepile.net/ (pick YAML for the language) or https://paste.debian.net/ (pick YAML for the language).
Please take the time now to review all of the rules and references in #rules.
You need to look in the log file for what comes before all that
Setup failed for dependencies: http, webhook
Those bits may shine more light on the problem
So it does say it fails on certificates and default config - I'll try to re-add them manually through cli and restart but that's the first time I've run into such issue... https://pastebin.com/y5UGGTWh
hi all.... i need help. i'm restoring my backup to a new installation of home assistant (deployed via the ova template) and i get all this errors
Check the pinned messages to see if the known issue is the one you're facing
And please don't share images of text, share text using a code share site
can't copy it from the esxi server, sorry
my issue is not the one pinned in the messages
please help me
...
Can any1 help me with manual certs, permissions and proper formatting.
fullchain.pem contains intermediate and host cert
privkey.pem contains priv key with bag attribues - or shall it just be key?
-rw-r--r-- 1 root root 2526 Mar 10 22:43 fullchain.pem
-rw-r--r-- 1 root root 3423 Mar 10 22:40 privkey.pem
lrwxrwxrwx 1 root root 4 Mar 10 21:42 ssl -> /ssl
I've been able to get to the safe mode gui - but cert I am using gave error about key value mismatch
2022-03-10 22:41:57 ERROR (SyncWorker_1) [homeassistant.components.http] Could not read SSL certificate from /ssl/fullchain.pem: [X509: KEY_VALUES_MISMATCH] key values mismatch (_ssl.c:4065)
Looks like your key isn't for the certificate in fullchain.pem
It is I jsut exported it from pkcs12 so I'm suspecting only formating issue
Here , try this:
http://letmegooglethat.com/?q=key+values+mismatch
Same with me
How are you creating them, and how are you getting the signed cert?
those are let'sencrypt certs created on pfsense with acme package. It is the cert which I have been using for last 2 months.
Normally they should be reloaded from docker config or backup but since yesterday supervised installation was failing. I made my biggest mistake when yesterday my supervisor remained disconnected after 3 upgrade failures from 2022.2.8 -> 2022.3.3 and I've decided to download backup and redo installation 🙂
never the less - what had happened with full backups restoration when they are missing certs - or maybe is it issue with importing them :/
I'd assume there's something going wrong with the export from pfsense
@bleak roost nothing really happens. You just wait.
I have waited all last night, today morning no lovelace still
Reboot and see how it goes
No that's not what I meant - I've removed intermediate from the chain and it is fine. seems I've mixed order. I'm asking why doing a full backup restore - does not re-create cert files. I've finally restored my backup, but core couldn't start because it was completely missing fullchain.pem and privkey.pem files - even after backup has been succesfully restored.
Did you reboot the whole thing?
yep
Then I would just reflash. Don't do a restore from the onboarding page. Set it up with a new user. Restore after it's up and running
Messing around trying to troubleshoot an install that never really got going is a waste of time
Ok, will try that and keep u posted. Thanks
@empty turtle cert files are in /ssl. Does that directory exist in your backup?
it should, I've made a full backup prior to yestrday attempts. I've been able to restore backup, manually scp cert files to /ssl and restart core which fixed that. Now I'm re-trying the upgrade to 3.3 - thanks!
So I fell victim to the "unhealthy supervisor" update. I dont have access to my SSL enabled on my HA setup and I don't have a MMK connected to it so I have not been able to flush the tables. Is the MMK my only option to clear the DNS cache?
So I screwed myself with jumping the gun on the "unhealthy supervisor" update. I just migrated two days ago and assumed that was the problem and decided to restart. After reinstalling home assistant I can't upload my old backups (error 400 bad request, in logs - can't validate backup). This is the same backup I used two days ago! Using a backup from earlier today ends in the system poping out lots of errors and stucking on "waiting until HA is ready". Has anyone else had any luck?
I also tried to downgrade to supervisor 2022.3.2 (ha supervisor down --version 2022.3.2) with no luck.
Same
Right now doing a clean install of the latest version, installing all the modules and then doing a restore. Hope that leads to somewhere functional 😉
I wish you luck! Do you also get the error 400 when trying to upload old backups?
I cant even get to the supervisor/backup page, I just get "Could not load the Supervisor panel"
Re-flash your microSD and do it again with a different approach
Like @crude inlet said "Then I would just reflash. Don't do a restore from the onboarding page. Set it up with a new user. Restore after it's up and running
Messing around trying to troubleshoot an install that never really got going is a waste of time"
The worst part is that I had my old system on another sd card that I managed to destroy today... while trying to back it up '^^
it happens, for sure there would be a solution, in the meantime you just have to trial and error
I still suspect the problem with restoring from backups has it's rot in todays unhealty supervisor problem.
Since the error of validating backups started poping up today as wells, and only backups prior to today gets the error 400 message
@tawny helm that issue is just a matter of waiting. Avoid tearing installs apart and reboot every once in a awhile until it is gone
The problem is that I already jumped the gun and assumed the problem was me. So I already made a clean install, but now I'm unable to restore from my backups.
Try a few of the backups
@crude inlet and @tawny helm for me the @crude inlet suggestion worked
I tried plenty of backups, most of them won't even upload. Tried restoring from a new user, and getting stuck in waiting for HA. I'll give it another shot now, trying to restore as little as possible and see what happens 🙂
So are you up and running now? 🙂
yep, back on track
👍
In the same boat
Did you use encryption on your backups?
nop
That might also be what's screwing me over now 😫
I have some backups for an old Rpi 2 on the garage
Did you do a full restore? And did you reinstall your old add-ons before restoring?
was an old one from there that worked
Installed fresh the most recent version of HA, afterward installed the components that i have on the backup... nodred, samba, all of them.
After that was trial and error with restoring different full backups to see what worked
Welp guess I'm going to have to try that route
The issue you were having after restoring the backup was that HA wasn't being able to be reached when you put in the local IP?
correct
Mmm okay once you added all the addons to the new install your most recent backup worked?
Been at this since 9pm last night only had 2 hours of sleep trying to figure this out
one of the oldest
Dont, take a break, sometimes our brain does not get out of the roundabout, for me a good night sleep and a walk do wonders
Oh no, my oldest is 2/28
Some success here I manges to restore ssh terminal, but z-wave JS doesn't work.
I'll continue tomorrow, hopefully some real fix of restoring from backups have emerged.
what is the uboot mmc commands to boot HA OS partitions? i have only booted to files from fat partitions before. 😦
After close to 24 hours creating a new instance and loading addons before restoring has brought back HA
So I just backed up a pi3 wiped the card and went to migrate to a pi4 and the restore never works nor does the pi come back. Am I missing something obvious? Can I not do a full restore?
not sure if this is the right channel, but anyone here have HA in a VM on their UnRaid machine - and run both a zigbee stick and zwave stick? I'm running into an issue where they both show up as the same exact usb device name
Bus 002 Device 011: ID 10c4:ea60 Cygnal Integrated Products, Inc. CP2102/CP2109 UART Bridge Controller [CP210x family]```
Check /dev/serial/by-id/ and see if that helps @jolly talon
Hi! Glad y'all have a Discord server as I am trying to back away from Facebook.
Curious how many of you have installed HA OS on an Intel Nuc? I read the docs but hoping to get a simpler install than all that
Easier than HassOS? Maybe you should define simpler. I use the docker image myself with docker-compose…..which I like….ymmv though
I ended up having to install an UnRaid plugin called 'USB Manager', and then that allowed me to assign both USBs to the VM even though they had the name name. so I think I'm good now!
More than 4600 people have (6.1% of installs). https://analytics.home-assistant.io/#installs Including me. What do you want to know?
Also Facebook, 🤮
Is there a way to get an older version of Home Assistant that doesn't have the restore bug so that I can restore a backup?
What restore bug? And yes you can install any version via the CLI, ha core update --version x.y.z
So you want to install a different supervisor version, not core version. Unfortunately the supervisor version will automatically update, you cant stop it. You are probably going to have to wait for the next release of the supervisor.
Damn
I can't get a restore working well even with the tips from mrzsilva and Leo's Gospel
good morning guys, have a short question, my homeassistant sdcard crashed - i have full daily backups, but my zigbee devices are completly missing ... after restoring the full backup - some hints ? or ideas.
Mosquitto broker MQTT with Zigbee2mqtt on cc2531 - i already did a partial restore with only theses applications but didned changed anything
Zigbee2MQTT uses the coordinator_backup.json and database.db for the devices
i copied the whole folder zigbee2mqtt , in the zigbee2mqtt are all devices , but i dont see them in the mqtt broker
has the Mosquitto broker also a seperate database ? there i see only 9 devices and in the zigbee2mqtt are all devices, but in the dashboard all entitys are shown as not available
Nope, the broker is just a message passer
so the folder is the right one ? \homeassistant\config\zigbee2mqtt - stopped the addons and copied the files over
how can I downgrade the supervisor?
/usr/share/hassio/homeassistant/homeassistant.log
ups
ha supervisor update --version 2022.03.1 doesnt work
I need to resore an backup .. but HA 2022.03.02 and 2022.03.03 wre not working with resore backups
@noble beacon posted a code wall, it is moved here --> https://hastebin.com/idoxigojew
Hi, im trying to restore my backup on a new HA OS VM but add-ons won't get installed. Error (one of the addons, all errors looks the same):
@noble beacon Rule #6: Please do not post codewalls (text longer than 15 lines) - use sites such as https://www.codepile.net/ (pick YAML for the language) or https://paste.debian.net/ (pick YAML for the language).
Please take the time now to review all of the rules and references in #rules.
The fix is here: https://github.com/home-assistant/supervisor/pull/3508. Just not sure when it will be released. I might try hot patching that later as I’m in the same boat.
hmm ok, thanks
Wasn't a codewall, mate.
Yeah, I have 100 zigbee lights and like 50 zwave switches not working right now because I tried to upgrade to a pi4. Piss poor timing due to that bug. I’ve been trying to add all the add ons and then restoring with very limited success.
Pretty rekt, damn... Who's to blame for this?
Not testing code before pushing it into release :/
Good news, you're entitled to a full refund 😛
thanks, what's the alternative product?
Mistakes happen often, this is the first time I was able to grab some motivation to report it here
Is there an easy way to patch that file from file editor or something?
No
Ok, that is what I figured but wanted to check. Yeah, given you can't downgrade the supervisor and even older OS versions upgrade it there isn't much I can do. Its either rebuild completely or wait for the patch.
Since yesterday I have had trouble with zigbee devices, today I tryed to install fresh and use a backup...bad timeing... i need bitwarden for my work ... I hope for a fast hotfix
The HomeAssistant WAF is broken at the moment
Where does the supervisor beta pull from? I saw they just merged the fix for the updates into main. I’m just not sure when the next release is or if the supervisor beta pulls from main.
That would say Join Beta Channel or similar
But there needs to be a release to update to
I guess the release is manual?
somebody has to do it, yes
I'm struggling with missing images constantly. ZwaveJS only seems to work for a while after a fresh install then I get spammed with
Image homeassistant/amd64-addon-zwave_js:0.1.54 does not exist for addon_core_zwave_js
type messages in the log, I don't think I've been able to perform an upgrade all year, they always fail with a missing image,
ghcr.io/home-assistant/qemux86-64-homeassistant:2022.3.3: No such image: ghcr.io/home-assistant/qemux86-64-homeassistant:2022.3.3
but usually deleting the VM and doing a fresh install will work? maybe I have some bad internet settings but all other integrations seem to be working 🤔
this is a supervised install btw
ah looks like things have moved to debian 11, let me go upgrade all my tooling and see if it fixes it
Thanks for the morale and thoughtful support yesterday! I finally back up and running again! The biggest problem appered to be duckdns, both the addon and in the config.yaml 🙂
Your welcome, it world is done of these things
And this is why I do a docker container install on my pi4
Now my HA is back... fresh install, changed to the dev channel, restore by backup und changed back to the stabel channel
Beta channel? Or is there a dev proper?
Beta is at the moment the same like stable - I think, it only works with dev
How do you change to dev?
ha core update
after reboot I restore my backup successfull
after that I used. ha supervisor options --channel stable ... ha supervisor reload .... ha core update
I'm following this guide to get HA on my raspi 3. https://www.home-assistant.io/installation/raspberrypi/
But nothing comes up when I search for https://homeassistant.local:8123/ or https://homeassistant:8123/
Keep getting "This site can't be reached"
Same error
Are you running a clean instal?
Yeah followed the steps exactly
Try the LAN IP
Is it connected to your network?
via ethernet yes.
Check your router for ip address
I don't know my raspi IP address?
If you can connect a keyboard and monitor it's displayed on the console
Forgot about the router... one sec
You can also download the app fing and it will ping all online devices on your network and give you the IP address
Yep, manual ip address worked. I'm assuming I'll have to use this address every time?
If you can find out why mDNS isn't working, and fix that, then you can use homeassistant.local, otherwise it depends on whether you want to run a local DNS server or not
I'm brand new to all this so I'm not sure.
remotely meaning while away from home wifi?
Yes
Yeah I'd like to
As was I a few years but you learn, if you need help feel free to reach out
Then you'll probably end up being able to use the hostname you use for remote access locally too
Depends on whether your router supports NAT Reflection or not
This is the method I use to access HA remotely and on local network, you won't have to use the IP address after you set it up
Okay thanks, I'll have a look.
Assuming that NAT Reflection works at least
It doesn't like my VPN
What does that mean?
I had my VPN deactivated when I accessed HA, but now that my VPN is on, it won't connect
I can't even connect to my router
I typically have my VPN on all the time on my PC. I'll look into the tunnel
Easy method a… WireGuard
Or you use your own dns server like AdGuard for your home network and get a domain etc and port forward
So I was able to recover my backup and get mostly up and running thanks to @sterile inlet's tips.
My ZWave JS integration however changed USB ports and I have no idea how to recover that network.
I think the new USB is now AMA0 from AMC0, any tips on how to reload an recover that? I have the keys
Hi, Just tried to restore a full backup (using HA OS on ESX) after the restore the gui is not available at port 8123, the observer is available. I have access to the ha console, core logs show many errors about unable to setup, anyone can point me to the right direction?
Hey guys I have a Linux image of home Assistent os running as VM. Apparently ffmpeg is not installed on this version. Does anyone have any idea how to?
Hi everyone, yesterday i wanted to rollback to an earlier backup i crated of the system....when i restore nothing. on a screen i could see HA CLI and all looks normal but if i ran "core check" it gave me somthing that it could not load SSL keys.....was restore tested on latest version?
now i will try to restore from full backup with parts
uhh if i try to restore a part from back up....nothing happens...how can i check it restored?
Guys I´m trying to do a similar operation.. I do have an Ubuntu server instance of HA and I would like to get the configuration (backup and restore it) on a raspberry PI4 running HA OS.. of course the devices are different and have different mac address (IP address at the moment is different as well). In the past I did use the GUI restore procedure and it did work without any problems, from what I read you are experiencing issues with the new HA CLI.. and I guess I´m running the same version Zeze21 is using in VM on the PI4.. if you give me few minutes I will let you know what happen wile I restore the backup of the running server version (core-2022.3.3 that is same on the PI4), on the PI4.
Keep u posted..
if you use the samba share option you can check files time stamps in the configuration or other folders 😉
ok, will install samba
as i was already bussy, setup HA again on different SD card, installed Zwave js andzwave JS to MQTT......then did do a partial restore for those components....now that part is working
ok, installed samba. When i go to HA server with Samba from my windows, i only see in the backup directory the backup file i uploaded
in config dir i see standard clean HA files
I use macos and samba I can see all folders of HA and right now the restore of the config is in progress as soon it is finished I tell you what I got here ..
Ok
it seams the restore procedure did work I got the complete configuration on the new instance of course not everything is working because of the different IP address of the PI4 (device wise there is a different configuration) I would need to change the IP shutting down first the server currently running.. this might take a few minutes more but.. I think.. I´m pretty sure (restore full backup) works really well.
hmm weird
the procedure I did was
- restore the image with the PI immager
- started the PI and initial login.. installed the samba share
- backup full the running server
- move the backup to the PI (copy /past)
- restore the backup (the PI rebooted and now on both side all is there, integrations and configuration too)
i did use balenaEtcher....
1 installed fresh image of HA
started up HA, created user and logged on
uploaded image in backup
then i did restore
then nothing.....then i installed DUCKDNS and configured it
then i installed zwave JS and Zwave JS to MQTT
then i restored both zwave modules from backup and now zwave seems to work again
I will try full restore now...is that possible
cause i had automations and also Node red scripts still need to work again (and of course the zones and Users but that is the least i am worried about)
I think as I wrote above.. when the two instances are running in parallel.. it can create some conflict at network level.. this is why the devices are not working at the first time.. anyhow I didn´t restart yet the instance (and it is a little strange but the backup in my case worked fine as I wrote by just copy and past from the old server to the PI4) I got all zones and integrations (alexa, Tasmota, LocalTuya and so on) running correctly (login all works).
Probably the upload of the backup via GUI was the problem.
Would you tray to copy the old instance to the new PI image using samba?
ok, will do that...but in a while as i now need to pickup my kid...thanks for the help so far
mmm just a side note.. only the Add-ons looks having problems right now.. not all was configured but it isn´t that bad.. custom components are restored as part of the configuration folder.
Do you consider Z-wave JS to MQTT a custom component?
Hi guys, I got a Home Assistant blue and I'm stucked at the preparing home assitant page with the error:
[supervisor.docker] Can't execute command: 500 Server Error for http+docker://localhost/v1.40/images/create?tag=latest&fromImage=homeassistant%2Fodroid-n2-homeassistant: Internal Server Error ("Get "https://registry-1.docker.io/v2/": x509: certificate has expired or is not yet valid: current time 2021-02-03T04:53:38Z is before 2021-11-30T00:00:00Z")
21-02-03 04:53:45 WARNING (MainThread) [supervisor.jobs] 'Updater.fetch_data' blocked from execution, no supervisor internet connection
21-02-03 04:53:45 WARNING (MainThread) [supervisor.homeassistant.core] Error on Home Assistant installation. Retry in 30sec
After searching here and everywhere I tried several things, one of them the instructions followed here https://community.home-assistant.io/t/solution-to-a-hanged-preparing-home-installation-screen-during-new-install-on-a-raspberry-pi-4-dns-fix/273803 about the dns issue. Now I have the 1.1.1.1 as dns server but still got the problem.
It seems to me like a problem with ntp but I'm not able to fix it. Could any one help me with this?
To try to troubleshoot the above issue I wanted to install some tools, like ntp tools but I cannot find a way to install any package from the CLI. I saw it was supposed to be a debian distribution but apt iss not installed so i have no clue how can I install packages, any idea about it or on how to resolve the main issue described above?
Another thing that I'm seeing is that my docker network is 172.30.32.0/23 while my host ip is 192.168.50.150 (this last one is correct for my home network), could this be a reason for my problem?
I just discovered (by shutting down the old server and changing the IP address on the PI) there is a problem by restoring the configuration of MQTT.. working now on it
Okay.. I found a small work around.. MQTT didn´t startup correctly because customization was set to true in the config.. I set the customization: false, MQTT restarted and now seams everything okay so.. it looks there are issues reloading the MQTT from backup.. therefore it isn´t Z-Wave only but please check the MQTT Broker log 😉
hi everyone., i just finish ha installation as a vm in windows.. when i access the ipaddress:8123 it doesnt let me add a new account
strange.. did it complete the setup phase without errors?
it did not bring me to the setup as seen in youtube guide.. im using haos_ova-7.4.vdi
then i navigate to http://192.168.1.9:8123/ then it would redirect me to http://192.168.1.9:8123/auth/authorize? which as me to login
wow I did never try the VM version.. hopefully someone using it can be giving you some hint on it.. it sounds like the instance isn´t properly configured.. using the PI image was working really well for me and it started up after the first time requesting to setup the new user and password as it should.
i tried using pi before and it worked, it just that im using pi as pi-hole server
well I repeat.. I hope someone using the same config you wish to try can be more helpful I do not have knowledge on this setup environment..
Hello. Looking for help. I can't seem to start my HA after doing a restore. Error message I am seeing from supervisor is: Error: 'Supervisor.restart' blocked from execution, system is not running - CoreState.STARTUP
I'm sorry that I can't be helpful. But I have exactly the same issue.
Are you able to reach your HA over port 8123?
No. It says unable to connect to home assistant
Maybe we have the same problem?
I upgraded to the latest version of the home assistant OS and then tried to do a restore, issue happened after restore
For me the same. It looks like it's still bugged. There was a closed bug related to restoring backups. Maybe it will fix your and my issue. https://github.com/home-assistant/supervisor/issues/3501#event-6225265680
But it will take some time until the bugfix is in the release
Thanks. I'm not really sure where to go next
Take a rest. I'm already trying to restore my HA since 48 hours. It just doesn't seem to work right now.
Have you seen the comments in #genera on a similar topic?
You are right, there is a similar topic. Maybe let's try to follow up there.
I've not used discord before today. Shall we chat over there instead?
i have the same issue when i do from CLI command HA supervisor restart
if i do "ha core restart" it says "unknown error, see supervisor"
Me too
What happened before you had the issue?
i am now doing "ha core check" and is processing
I upgraded the OS, had some issues, tried to restore and really stuggling now. There is also a chat about this in #general-archived
but i am a noob on this by the way
you and me both 🙂
for me it is shitty cause my heating system is depening on HA + node red
do you have SSL installed cause the check gave me back a error on SSL
I think so. There is chat about this in #general-archived - shall we continue there?
yes, will check...the more ppl checking the sooner it gets resolved
Hey, i just downloaded HA to my raspberry pi, how do i know if it is working?
In the browser of your Desktop system, try homeassistant.local:8123.
it just says this website is unavaliable
Sorry I am no expert. The installation instructions are here if it helps: https://www.home-assistant.io/installation/
@visual axle presumably you have a screen attached to the pi?
To checked it's actually booted and all
homeassistant.local doesn't always work depending on your network. Finding the IP of the Pi might help
Can you have two instances of HA on the same network. I'm just installing a second as I have issues with the first
You can have as many as you want
But out of the box they will all have the same hostname
So you won't be able to go to homeassistant.local anymore
Would need to use the IP
Depending on the way you install HA you can change that accordingly
Or change the hostname on one of them from the Supervisor > System > Host card
Ok. Any ideas on the questions below about problems with HA after doing a restore?
above sorry not below!
Do you have the IP of your installation?
Maybe you were restoring SSL certificates? And are still trying to use http:// ?
oh sorry i fixed it, i was totally not using a pi 1
I did a complete restore. Yes I do have the IP address
Have you tried using the IP?
Still can't access it?
Are you physically connected to it that way you're seeing the supervision warnings?
I can connect but it won't load. Yes I can ssh to it to see the su warnings
hello im having trouble installing home assistant on my windows 10 pc i get to put in the ip address and then when i pull up on browser this is the error message[supervisor.homeassistant.core] Error on Home Assistant installation. Retry in 30sec an its loops and wont get pass this
hey guys, im wondering if i would install Home ass on my PC via VM would my Built in Bluetooth be useable for Home assistant BLE Tracking?
If you pass the Bluetooth adapter through to the VM
great
Are you able to update to 2022.3.4 - I am on hasos and can’t update, even on commandline it won’t work. And release is 18hours old, should be already there right?
If you can SSH then everything else is running. It's only HA that has a problem
what about ha core logs
are you sure it's not because you had SSL before?
Presumably this is on a VM? You used ethernet as a bridged connection?
the build failed
Ah, so there is no 2022.3.4 atm 🙏
Doesn't seem like it
only way you'd have known is from github as well
it never made it anywhere
It's dead and starting to decompose
I have a file of the ha core logs output. can I share that here?
I did have SSL before but I still have that
so you're trying to access ha using https://your_ip:8123 ?
you can use pastebin or any other such website for the logs
Just tryig to work out how t use pastebin. it starts like this: [31m2022-03-12 19:55:50 ERROR (MainThread) [homeassistant.setup] Setup failed for diagnostics: (DependencyError(...), 'Could not setup dependencies: http')[0m
[31m2022-03-12 19:55:50 ERROR (MainThread) [homeassistant.setup] Unable to set up dependencies of websocket_api. Setup failed for dependencies: http[0m
[31m2022-03-12 19:55:50 ERROR (MainThread) [homeassistant.setup] Setup failed for websocket_api: (DependencyError(...), 'Could not setup dependencies: http')[0m
[31m2022-03-12 19:55:50 ERROR (MainThread) [homeassistant.setup] Unable to set up dependencies of cloud. Setup failed for dependencies: http, webhook[0m
Yes. Neither https or http work
yeah you have some serious problems there
not even safe mode could kick in Setup failed for safe_mode:
try ha core rebuild
Error: Unknown error, see supervisor
I was chatting with some other users earlier. about four of us have issues after trying to do a restore
ok
Anything else you think I could try?
you have a backup safe somewhere?
ok good so doesn't mater if we nuke this one
correct
from ssh remove everything in /config and reboot the host
Ok have done that and am rebooting
is it possible to get my supervisor back from .dev1102 to .03?
System is up. What do I need to do next?
Hi @high token can you please advise what I need to do now?
Version
supervisor-2022.03.dev1102
Newest Version
supervisor-2022.03.3
@sterile inlet you need to change channel back to stable
then reload supervisor
core is stable core-2022.3.3
and it should list .03 as the newest version
So did now HA load as a fresh installation?
ha supervisor reload .. .nothing
Yes 🙂
Actually no. It looks like some stuff is there
maybe you didn't actually delete everything?
as long as configuration.yaml is new the rest aren't being used
I'm not sure what was wrong with your backup before
one option would be to partially restore only your configuration from it
ok thanks. what would you recommend me doing now?
upload your back up. Restore - select partial and only check HA configuration
if you end up back in the same place, then unzip your backup and take everything from /config in that backup and manually move them into /config overriding whatever was there previously
Ok if i try and do a partial restore I don't see a config folder
I get options of Home Assisant and then folders: local add-ons, Media, SSL, and Share
you want the first one that says "Home Assistant" with a version
is it impossible to do: ha supervisor update --version 2022.3.03
I missed the PR on that. Just saw it
in the actual tar backup file it's called homeassistant.tgz or similar
I'm not sure about that anymore. Generally you cannot downgrade Supervisor. But you should be able to go "back" to whatever is latest if you are changing channels, or so I thought
Ok so I have backups in the backups folder, but I can't see them under supervisor
The list is empty now. HOw do I get a screenshot on here?
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.
I'm having the same issues that @solid furnace is having. I recently tried to restore ha from a clean install through back up. Now I'm getting all kind of dependency errors. Any tips? Tried searching the community forums I could not find a thread that was specific to this issue I see here.
Is it possible to install packages to HAOS? I need to set up nps for NAT traversal (I don't have public IP address with my Fiber provider, but I do have a VPS with public IP and nps installed). I used HACore on Debian before but I wanted to use Supervisor so I switched to HAOS. Seems like it's locked down and you don't even have a normal Linux shell?
Creating an addon is the way to do that, or find one that someone else has built
Thanks. I looked up and no one has created addon for it. I have never tried to create my own addon as it seems not for beginner. I'll read the relevant documentation and maybe try it.
It's not that hard for simple things. https://developers.home-assistant.io/docs/add-ons/
Yep I am looking at that one. I know how to make it work directly in Linux box but Docker is the thing that scares me 😂
Maybe use tailscale?
That’s an easy thing
I actually only want to forward port 8123, thus I don't want to use a VPN
Hi I have a question I just installed home assistant and I'm unable to find out the addons tab
Hello. I'm migrating from my ODROID-XU4 to my new ODROID-N2+. I did a full backup on my XU4. Installed the newest fresh image to my N2+. At the first mask (user account creation) I choose the restore option. The backup file is uploaded, but that the N2+ is not accessable anymore. I was waiting a long time, nothing 😕 I'm using a self-signed certificate for SSL. I use the same IP address than before. Any ideas? The process is so straight forward, but didn't work.
You installed HomeAssistantOS?
I believe there is no AddOn functionality if you installed with Python venv or Docker since the AddOn is provided by the Supervisor which only exists in HAOS or supervised installation
@brazen jungle same issue. It won't even load from the front end. I've tried. 3 different back ups from different time period over months. same issue. Not sure should I just start over?
Did you try it via command line. Maybe this is an option...
Is the Backup Restore Feature really broken?
I will try it now via CMD: ha backups restore <slugname>...
Restore via command line is also not working 😦
I'm trying that now @brazen jungle if this doesn't work. I'm going to attempt to roll baby to an earlier version of home assistant and supervisor, and try to rollback from there. Something seems up with this version(2022.03)
@brazen jungle hi, im running ha core on a docker, and bind the local volume to the /config path, im a little confused about the relationship between the .storage folder and configuration.yaml file, can you explain it for me?
The .storage/ folder is in the same folder as configuration.yaml
That's about where the relationship ends though
.storage/ is used by the things you do in the UI. configuration.yaml is entirely manually controlled by you
i search for the tutorials on google, we can integrate devices by configuration.yaml files, but if i integrate a device from the ui, i can only see the modification in .storage folder, this is where i am confused
Right, that's correct
If you configure something in the UI then that changes things in .storage/ - other than scripts, automations, and scenes
Keep in mind things that you find via Google may also be horribly out of date and giving you incorrect information
and should never ever edit anything in .storage of course
so can i totally remove the /storage folder, and only use configuration.yaml file, in this way i can controll all my data?

Sure, in the same way that if you don't use the walls of your house you can remove them, and totally keep your house
In short - no, if you delete that folder you break HA
I'm not sure why you think you're not in control of all your data though
i mean if i modify my data from the ui, then it will store in the .storage folder, which i cannot edit the files in it😂
But, you're not supposed to - you're supposed to use the UI to change that
If you want something that's 100% manually configured, then the 1970s are calling
im supposing this scenario, first i was using the ui to manage my devices, and lately i want to migrate to using configuration files to manage them.
will HA combine these two parts of data?
just wondering about the data behavior of HA
Mostly it's about following the docs - if something is UI configured then it's UI configured, if something is configured in YAML you use YAML
so modify the device info which define in yaml from ui, HA will write it back to yaml files?
What?

Are you a time traveller from back when HA didn't have UI configuration for things and want to go back to those days?
The only way to "modify the device info" like that would be to re-write HA
I still don't understand why you're so keen to do it all in YAML
Are you saying that you can't handle the mix of UI and YAML?
I don't see that changing in HA any time soon - like in the next many years
Sure, there's a move towards more UI options, but some things look to be remaining YAML forever
Maybe you'll be happier paying for a commercial platform that's 100% UI managed
Even OpenHAB is a mix of UI and config files
I'm trying to get running with HA OS on mac via virtualbox. Issue I'm having is that there are no IPV4 addresses being assigned to the VM. Has anyone else had this issue and resolved it?
Any successful able to login into home assistant via port 8123, after a fresh install and restore of a backup from version (2022.03) ??
Guys, I do have an old x86 32bit.. currently installed Ubuntu and I thought to change to Debian buster as per HA works better on it. I did see a guide for the installation on Intel NUC and read there that it might be possible to use HAOS on 64bit systems only (do I´m correct here?) therefore if I install Debian 32 bit on the system I can run HA only on docker image if I understand correct... can someone confirm if I´m understanding correctly please?
Been directed to ask this in here :
Got a weird problem which I can't find an answer to ... I've got Home Assistant installed in a VM in Proxmox. The VM has 4Gb of RAM but if I ssh into 22222 free -h only shows 905Mi. I've found something about this on an RPI from a while back but I don't think that applies. This VM used to work fine and show 4Gb in the OS but I migrated it to another Proxmox host in the cluster this morning and now its doing this. Anyone got any pointers ?
HAOS would replace Debian
I'm not sure there's any support for HAOS on 32 bit, I'm not sure there's even support for Docker on 32 bit
on Ubuntu server 18 it works under Docker.. but I do have some limits of course there. I just try it on a PI4 and HA OS is what I really would like to have on the "server" that currently is running on the 32bit machine. I know that the PI4 can run Debian 64bit for sure.. but I´m a little concern about the stability in long therms with the PI4.. altenativelly I could mount an SSD on the PI.. and dismiss the server at all.
but honestly I would try both installation methods and see whatever.. it works or not (and of course take a good backup of what is now working for sure 🙂
How do you expose docker to your network? I have a WSL (Linux) and I have installed Docker. I pulled HA image.
You have to use Host Mode networking, which isn't supported in Windows
I am technically on Linux xD
Well, yes, but no
Ah shooks!
hey good day, how long does it take on the Preparing Home Assistant screen?
What would you suggest me to do then? @humble mirage Shall I split boot my computer to have Ubuntu
Hi. I keep getting this error, after updating to newest edition. Downgrading didn't help. ANY idea what to look for? https://pastebin.com/drwLx4wC
2022-03-13 23:28:51 ERROR (MainThread) [homeassistant.config] Unknown error validating config for template platform for sensor component with PLATFORM_SCHEMA
I'm just retiring my Pi and building HAOS in a VM. I logged in, uploaded my full backup into the backup tab and triggered a restore. How long does that normally take?
So many factors at play. How long have you been waiting?
5 or 10 minutes
Just for the backup? Or to get the image and the backup up?
I clicked yes to "Do you really want to wipe your whole system?". It's just sitting saying "Lost connection, reconnecting"
I'm assuming it needs to download all my integrations and add-ons
or isn't backup that smart?
It's doing something, using about 7.5% CPU according to VMWare
That’s good, just be patient.
The system is fairly basic. 4 smart plugs and 20-odd Zigbee devices. How long would you wait before deciding it was broken?
I mean I would have gotten the image up,and then once up, restored the backup. I don’t run hassos though.
Isn't that what I did?
It’s not 100% clear….but if it is…..then it should take that long to reload your backup
if you made it into the Supervisor then everything was up
monitoring the supervisor logs will tell you what's happening
The console is still running in VMWare and I have a HA> prompt. The web UI is not loading
It did
are you using https now then?
I think it's failed. I have errors that core_configurator, nodered can't be restored and that 172.30.32.1 isn't responding
Am I hearing the subtext that the backup didn't include all the add-ons, blueprints, integrations and the HTTPS cert?
if it was a full backup it should include everything
I was asking because if you are restoring ssl you should change to https
a lot of people still try to use http and forget they were restoring SSL
Thanks for your help. I think I'll go back to the old system and try again in the morning. I need to get to bed. Work in the AM
I think I just foudn part of the problem. I was using the Bluetooth tracker and the new machien doesn't have Bluetooth
Why does home assistant docker container need to be privileged?
And, I suppose while I am at it, why does it need host networking 🙂
Hmmm, I’d guess the host networking is about seeing various discovery protocols
Are we able to restore backups now?
Either have the computer run HAOS directly, run HAOS in a VM, or install Debian and then Docker for an HA Container install
Does anyone know why I'm flashing my eMMC on IO board get's stuck at 41% for hours!
I run rpiboot, set the boot jumper, connect the device, it finds the mass storage Compute Module, use the balena imager, start flashing, then it get stuck at 41% for how long I leave it, last night I left it for 5 hours !
More than likely issues with the eMMC. How old is it?
new, CM4 32GB 2GBRAM
How many times have you tried? If multiple, does it get stuck at the same place every time. If not multiple, try starting over with a new image download.
Yay found the problem ... needed to Disable memory Hotpluging in the VM
Nope
Debian is the only supported one? I’m familiar with Ubuntu
Ubuntu is ok for HA Container, if that's the way you roll
Ubuntu isn't as good a choice as Debian for servers, but mostly it's not terrible... except when they push a broken package
Anybody have tried to install Homeassistant os on intel NUC celeron N33xx ?? Is it run natively or vm?
Or because Snap comes pre-installed 🥴
So.. in the end I did try Debian 11 32bit and of course the procedure to install HA isn´t working .. anyhow I found a procedure to install docker under ubuntu 32 bit (that worked also under the Debian 11 32bit) once installed docker and via docker it was possible to get the HA instance (not supervised because, probably the system was already smashed up with other install attempt). Therefore.. yep.. 32bit wise the machines can run HA. Now I´m doing another attempt with another install procedure that involve the usage of the HA OS x86 64bit. Will keep you posted as soon I get a clear method to get it run (hopefully as it should) 😉
For an HA Core install on WSL, yes
I agree with @humble mirage.. Ubuntu 18 worked quite well with HA and honestly I´m having a big deal of troubles with Debian on my old "server".. anyhow the advance of the Debian version (Home Assistant OS) is that somehow we can run a little more thinks on it
I am having issues installing the prerequisites
Then the Warning box at that link is relevant
Err:1 http://deb.debian.org/debian buster/main amd64 python3.7-venv amd64 3.7.3-2+deb10u2
404 Not Found [IP: 199.232.138.132 80]
E: Failed to fetch http://deb.debian.org/debian/pool/main/p/python3.7/python3.7-venv_3.7.3-2+deb10u2_amd64.deb 404 Not Found [IP: 199.232.138.132 80]
E: Unable to fetch some archives, maybe run apt-get update or try with --fix-missing?
please consider one of the other installation methods
Python 3.7 ... eh, no
You need 3.9 or 3.10
You also need to update to Bullseye
As an alternative, Home Assistant OS can be installed in a Linux guest VM
Install Debian Bullseye, that has Python 3.9
Or, install HAOS in a VM
I would prefer to use normal WSL install
Then you get to learn a lot of Linux
I don't have a VM
of course it is 🙂
I am running WSL
You can have one though... it's not exactly hard
Sure, but #installation-archived message
how can I upgrade python in WSL?
Do you want to learn?
of course
Then can I suggest you try Google?
Here , try this:
http://letmegooglethat.com/?q=debian+buster+to+bullseye+upgrade
Keep in mind that in December support for Python 3.9 will be dropped and you'll need to install Python 3.10 from source and then rebuild your venv
I am on 20.04
Are you saying that if I run Debian, I can run HA OS. Since I’m running container (Docker) on Ubuntu
Indeed it depends from the hardware you use.. on my case Ubundu 18 was working with HA supervisor without any big issues. With Debian I do have some issues installing it as described in the official procedures. But I can confirm that Debian 11 64bit is the HA base OS.
P.s. Docker instance is what I got too under Debian 32bit but isn’t supervised
Okay good to know
what issues did you run into?
Docker CE isn’t running on the 32bit versions.. my machine is evidently too old.. the funny thing is.. On the PI4 I did successfully install HA OS and there everything runs.. I think to order a DELL Poweredge server 😉 there is possible to have a 64 bit OS and then try with it
ahh ok
@humble mirage mentioned there may be some issues with restore on current version. I've been really struggling trying to get my instance back up and running after having to recreate my Proxmox VE instance. I though I would be ok with a full backup, but it has been VERY wonky. Not sure what the best method to proceed. Should I roll back to an earlier HA version and try to restore and then start updates from there? Would that prevent the current known issues?
I was only able to restore a back up that was created last month(Feb 2022). Any backup that i had that was created in March 2022. Ended up hosed with a bunch of different errors. So I ended restoring from Feb 2022. And then (Unzipping/tarring the March Backup file, and piece by piece restoring the missing pieces that occured since febuary. #Painful
interesting, I thought going further back was hurting, but maybe it's better. Can you restore Feb backup into a current version of HA?
Thats what i did, backup-restore in the most recent version of HA..
I'm pulling my hairout trying to figure this out - I have installed HA countless times before without a problem but no matter if I try OS version 7.3/7.4 or 7.5 on PI4 I get this:
https://imgur.com/a/iZE4qRo
this is all 64bit - I tried 7.4 right now and same dockerd error
Have you got a working shell prompt?
my full backups from feb are kinda big. 27GB smallest one.....I wonder if I should pull the non-essentials out. do you know if size matters 🙄 , lol, but really.
Can you type commands to the host to ask Docker what it thinks is going on?
I'm unsure. I've never plugged a monitor into my Pi
anyone have experience with HA OS on a rp3ib+ & booting from a SSD?
what have you done?
Overnight I'm getting unable to connect to home assistant in web/app, both locally and elsewhere. Hooked up my rasp to monitor, and it seems like it's booting just fine. Collecting IP address and working fine. Unfortunately I didn't install SSH or samba (Silly). So is it possible to do that via CLI? Or any other ideas?
@daring lance if you have access to the cli you can check the logs
And yeah you could install the ssh addon but you wouldn't be able to configure it really
But you have access to what it would provide
Sorry for my lack of insight. I've really tried googling how to use the cli and using the help command without any good luck. How do I access the logs with cli?
Finally managed to get HA to install on my Pi4 but now when I try and restore an old back up it says: https://imgur.com/a/5xOrF8K
This was working less that 24 hours ago so no hardware has changed etc
not long after that it says "Failed to start Journal Service"
Is it possible to install home assistant on a proxmox server and use the audio jack to output music from it?
Tips on Remote Access to Home Assist with SSL and Dynamic IP to control anywhere
It may take a week to restore. Lol. But I say try it. If that doesn't work pull out the fluff. I've heard the database if you still are using the sql lite version can get really big, and can inflate the backup size
Hey, anyone on linux (ubuntu) and have a working SSL Cert autoamtic way? I thought of using Certbot. Trying to create a letsencrypt ssl but fail.
Challenge failed for domain x.duckdns.org
dns-01 challenge for x.duckdns.org
ha core logs for HA logs
ha su logs for Supervisor logs
@placid sage are you using the duckdns add-on? You don't need anything else
I am on Ubuntu, HA Core not OS. I did not install supervisor
wow, this is interesting....so, it did seem to restore pretty well with a feb file.....im assuming u were able to update everything to current version once it was restored??
Yes sir!!
Awesome. Going to give that a shot now. Thanks so much!!
So, I installed Bulleye on WSL
and also python 3.10
python3.10 -m venv .
Traceback (most recent call last):
File "/usr/local/lib/python3.10/subprocess.py", line 69, in <module>
import msvcrt
ModuleNotFoundError: No module named 'msvcrt'
what's wrong here?
P.S. what's the unix password for the user homeassistant that I created following this link:
the above command works with python3.9
but not with python 3.10...
You didn't set a password
so?
So you didn't set one... there's no password
what's the problem with python3.10 venv?
I tried with empty password and doesn't work
Google for
ModuleNotFoundError: No module named 'msvcrt'
I did
but no solution found
One question
pls
when the WSL installation of HA will be ready
how can I access files for configuration?
Shall I modify them directly inside WSL
or will they be visible also outside WSL in windows dir
?
tipically Samba Share is the way to access the HA files from outside.. but in a virtual machine I belive it works also drag and drop
Or not, as suits you
??
You can edit them however you want
also from windows?
As for the solution, there's many possible solutions there, you just need to try them and see what's going wrong
my problem is that the server is stored somewhere without a monitor
Or maybe it's time to accept that you don't have the skills required for WSL and use HAOS in a VM
🤣
you always reply to me, but value added is not a lot...

I'm trying to get you to use an install method you're capable of running
You're right though, I'll leave you to it
I'd like to learn but you don't seem to be willing to teach
Google is not always the reply
thanks for your time btw
hi guys. where should I ask questions about coinbee 2 installation errors.
thanks for pointing me in the right direction
If that's a custom component you'd want to ask in #integrations-archived
If that's ConBee then that's a #zigbee-archived question
can anyone tell me how I get my external drive to show up in media?
(running Home Assistant Operating System on a Raspberry Pi 4)
The docs show using /mnt/XXX in the path: https://www.home-assistant.io/integrations/media_source/#using-custom-or-additional-media-folders
As it says, you need to add the media_dirs to your core config: https://www.home-assistant.io/docs/configuration/basic/#media_dirs
But I’m not 100% certain this will work.
What are you trying to accomplish?
how can I find out the path that I need?
I have a lot off music (in mp3 files) which I want to be able to play. I can't copy everything to the py because its drive is to small. I have an external 2TB drive laying around, so I thought I can just use it.
Personally I’d recommend setting up a media source, such as a DLNA server or Kodi. https://www.home-assistant.io/integrations/#media-source
I know, but I don't have aditional hardware around to run it on and I'm a student so I basicaly have negative money at the moment 😅
Do it as an add on? There’s almost certainly one for it 😉
Configuration > Add-ones & Backups > System > Host > three vertical dots > Hardware
But that doesn’t show the mount path, you’ll likely need to ssh in to get that, if you can.
The only thing I found is the Plex media server and a lot of people having problems with mounting external drives.
I use Plex, but I host it on another machine so I can’t help you there. Realistically, HA isn’t meant to be a media server, so you’re rather barking up the wrong tree.
I probably already know the answer but worth checking with you lovely peeps.
Why is it that a 16gb sd card is quicker at onboarding than a 64gb sd card?
I’m thinking it’s either due to the smaller size or because my 64gb card is on its way out. #sluggish_HA_on_a_64gb_sdcard
Could be the speed of the SD card, could also be that it doesn't have to expand it
Is it possible that the card is on its wayout? Ha was so sluggish and took ages -eventually crashed and was a pain restoring it
but since I did a reinstall on a different (and smaller card) and restored using back up it has been refreshingly fast…
Quite possible, SD cards do have a limited life
May need to invest in ssd then….
I would go head and do it now. . . my sd finally gave out less than a year. SSD has been the way since, no regrets.
I couldn’t get a pi 3b+ to boot from an ssd.
are you looking for assistance at doing that? 😛
Naw….thanks though. I think it had to do with the brand/model. It had to do somewhat with the power sequence.
I just went with spinning rust and moved on.
it's not the easiest thing indeed
I would boot to the drive, get an error message, the unplug and plug the drive back in and it would boot fine.
I've seen this on issues on github
mostly on pi4 though because of the difference in its u-boot
Yes, I think I saw something about backfeeding power on the pi4.
Pi4 is quite power sensitive. Not enough power and can lead to issues
it really likes to have 3A
then you connected an external drive which also needs some of that power and can quickly lead to problems
that said I've been running on SSD on Pi4 for 2 years 😄
But a pi3b+?
yeah on pi3b+ I thought it was more straightforward
Also would be interested in brand. I don’t think it was a power issue when it comes to the power supply
I think it was more a timing of when the drive was powered up
compatibility with adapter is definitely an issue on pi4. But I was under the impression the same didn't apply on pi3 because of different way they do usb boot
is it possible to access configuration.yaml from CLI?
Yes
At least on a docker install, but pretty sure this spans all the install methods
You can access it on all, at most you need to install the SSH add-on
Unfortunately I did not install SSH. I really wanted to try to remove duck dns addon, or update it. I cannot find how to access config file though, any guidance?
What's the problem you're facing?
unable to connect to home assistant. Both with my pi running and without. Just happened overnight, backup gives same result
Yes, it collects the static IP address... And no, I did twice stupid not installing samba or ssh
Do you have a monitor and keyboard you can connect?
Yes, I'm checking the logs via CLI (Monitor and keyboard). There is some sort of network issue, but I cannot figure it out. I see some addons not getting their connections right
Do you use a wired or a WiFi connection?
Wired
Sharing details of the "some sort of network issue" may help others help you, though have you tried rebooting the Pi?
Sorry, I've been troubleshooting and trying stuff for a week. So many details lost in text. I see my ESPHome trying to collect the static IP to my ESP's and not getting a return answer. They are online (Checked in router)
All, I did some test with Debian 10 32bit and finally I found the way to install Docker-CE and run the setup of the Supervised without error according to procedures. The docker image is Hassio-Supervisor... and at the end of the docker install it says that in few moment (about 30 minutes passed) the CLI should start.. but nope. Installing HA core with docker on top results on the core only version that.. well it would it be okay.. to be honest.. and anyhow.. it looks to me a peaty that I changed from Ubuntu to Debian, Ubuntu for instance was a little more "flexible".. and Debian is a little too "Secured" giving a strange "sudo" limited experience. Anyhow as per someone else could be interested on it here the link to get Docker -CE on a 32bit system https://nagimov.me/post/running-docker-on-32-bit-hosts/ and indeed.. from the tests I´ve done it looks the only one really working. For the moment and anyhow for the future... well the 32bit will be (from my side) replace with a at list some newer-old server 😉
https://github.com/home-assistant/supervised-installer is the only way to run the trap that is Supervised, and 32 bit ain't invited
That said, just say no to Supervised
yes I read it but... on Ubuntu18.04 Bionic (32bit) it works.. without the full experience of Debian.. so as per we wana stay in spec.. HW wise.. HA better run on 64bit CUP´s
What do u mean.. you want to connect the ESP´s with the ESPHome integration? if yes it depends from the configuration you gave with the ESPHome add on (that isn´t the same think of the ESPHome integration) and by the way.. this looks more a topic for #automations-archived or #integrations-archived
anyone using traefik with HA and have x-forwarded-for working? I'm getting my gateway IP address as my header but I'm not sure if it's a docker issue or a traefik issue
I couldn't figure it out and think that NAT hairpinning was my issue
so I added an A record into my router's DNS for *.mydomain pointing to my HA host's IP address
I'm no expert, but... Is there a logical explanation for why wired connection does not work? I don't get a IP for some reason.
If I, for example, boot Noob OS, the wired connection works fine but not with Hassio. I have tried to manually activate the connection (CONFIG/network/my-network), via CLI and also via Supervisor but it refuses to work.
Any ideas?
What Pi is this?
When i boot Hassio, eth0 is not activated. I have tried to activated and set a static IP but no sucess. Tried the same thing via Supervisor but I same thing there.
if I login, nmcli con show. There is 2 ethernetslots
one i Supervisor eth0 and the other is my-network
I can only activate the Supervisor one
and if i do that, the connection to Home Assistant drops and i cant connect
I can see in CLI that the activated eathernet have the right static ip
I have tried to reinstall everything from the beginning but the same problem recurs
As said, If I boot Noob OS. The ethernetport works fine with the correct static IP
The ethernet port in Supervisor is default inactive. Activating it makes home assistant lost connection until I reboot and setup the WiFi instead
What version of Home Assistant OS?
latest, Home Assistant OS 7.4, Home Assistant 2022.3.5
Hmmm discords had a spring clean. Installation methods archived so all in here now?
There is no error message if i, through Supervisior activated wired and set a static IP. I can see in CLI that the connection is active and the static IP is there, but it dosent work after that
Everyone into the pool!
can ping the server after that
So I have a stupid question: If I install the NUC version of HAOS on a USFF PC like a NUC or a Thin terminal, will it have the Supervisor like on my Pi?
It would be the generic x86 image, but yes.
Thx
Only doing it cos I have the boxes gathering dust, and thinking they'll have a bit more poke than my 3b
HP T510 with a small SATA SSD. Dual core CPU, 4GB RAM shouldn't be too bad, and only runs 10Watts
I've just tried to restore the config from my old HA RPi onto my new Intel Docker build and it's gone really wrong. The UI is refusing my connect requests and the log in HA is spitting screens full of errors about missing dependencies, things like analytics, auth, http, person, websocket_api, webhook...
A number of people have reported issues with restores. but if you can use a code share site to share the start of the log I'm sure somebody can help
I don't know how to get the log off the box. I'm viewing it on the VM console in VMWare
All I have is an ha> prompt
The supervisor log says the core config is invalid
Is there any equivalent for "less" or "more" at the ha prompt please?
I have no idea, I don't use HAOS
ha core logs and ha su logs
That's not being useful as there is so much stuff it scrolls off the screen and I can only see the last screen full
in this case ha core logs is most useful
ahh
I guess you can't pull them off with samba
I haven't used ha os in a while (I am thinking I should just to help out) but I know you can drop to the shell by typing login
That gives me root on the VM but aren't the logs inside one of the Docker containers?
you can try tail -f /config/home-assistant.log
Should I run my HA on a docker container or create a virtual machine with the HA os installed on it ?
On my Synology Nas
Your choice
If you want add-ons then VM + HAOS
If you want native Docker then Docker
Are addons the only difference? I already had to apply a work around for my broker on my Synology Nas
Well, they're the main thing
The Supervisor also brings "backups", for those that don't know how to do backups
(they're really snapshots, but called backups)
Wel only expansion I'm planning to do is add ZigBee to the mix and since Synology DSM 7 does not support dongles anymore but the VM does it maybe wise to switch to VM
File doesn't exist
Or buy a network connected Zigbee coordinator
Tube sells some rather nice units
Okay only found the coordinator in USB dongles so far let me check
Yeah found it they don't ship to Belgium do
Pretty sure he does, folks in Europe have bought from him before
Belgium is not the only country in Europe 😋
You can always just ask him
There's a mail link on his site, or you can ask him directly over in #zigbee-archived
Got it thanks for the info much appreciated 👍 you really know loads about this stuff almost each channel I ask a question it's you responding 🤣🤣
hold on try looking in /mnt/data/supervisor/config
that should be the location of the config folder
is your sa docker install?
HAOS in VMWare
ah
@sage gulch /mnt/data/supervisor/homeassistant/home-assistant.log
the folder it's called homeassistant in the OS level. It's config as a symlink or something like that everywhere else
What do you guys think about installing Home Assistant on my old Raspberry Pi 2?
I think you will have a bad time.
There is no official image for this and it will be maddeningly slow.
Perhaps I could install it on RaspberryPi OS using the supervised instructions?
Wait. I lied. It is an RPi 3 Model B V1.2
There's a reason why there's a warning banner on supervised. Pay attention to it. You're better off with the container or hassOS anyway
Raspberry Pi OS is not officially supported, you will have to use Debian for Raspberry Pi 2
Well it’s an RPi 3 anyway. I was confused because of the generational naming of the RPis.
I have a Model B anyway.
ahh then haOS is your best bet
I am not afraid of tinkering. I kinda prefer it sometimes 😃
you choose the installation method for you, I personally run supervised
Got permission denied while trying to connect to the Docker daemon socket at unix:///var/run/docker.sock: Get "http://%2Fvar%2Frun%2Fdocker.sock/v1.24/containers/hassio_cli/json": dial unix /var/run/docker.sock: connect: permission denied getting this error when trying to type ha in cmd in a raspberry pi 4 (NVM FIXED IT WITH SUDO)
Followed all instructions in Home assistant supervised
Cant use the HA OS because I need to run a few discord bots and PiHole at once with it
and http://homeassistant.local:8123/ this link shows
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 4.154/4.154/4.154/0.000 ms
Device "eth0
eth0" does not exist.
[info] Install supervisor Docker container
[info] Install supervisor startup scripts
[info] Install AppArmor scripts
[info] Start Home Assistant Supervised
[info] Installing the 'ha' cli
[info] Within a few minutes you will be able to reach Home Assistant at:
[info] http://homeassistant.local:8123 or using the IP address of your
[info] machine: http://:8123
``` Looks like it is on after using sudo dpkg -i homeassistant-supervised.deb
Its installed but i cant connect sudo ha works
mmmm interesting I´m somehow in the same condition.. I did the installation of debian 10 (I know now we would need debian 11 instead but I had to be creative as per my cpu is 32bit) is on the "server" and successfully installed HA supervised, it start but stop exactly there.. so if it is a SUDO user issue (I suppose docker need some privilege).. mmmm ... let´s see
have a look to this.. https://docs.docker.com/engine/install/linux-postinstall/
This means that you're running an unsupported install, the Supervisor will complain at you and at some point things may break
You can use the AdGuard add-on instead of PiHole, and build your own add-ons for your bots if they don't already exist
Still, if you like knowing that your HA system may fail on you after you've ignored warnings... go for it
yep.. "Error: No such container: hassio_cli" after cleaning up docker installation and adding the sudo rights to docker... I spent too much time on this so.. let´s stop here. I guess that I will try later on a intel XENON x3470 system.. my goal actually is to run the Supervised on the "supported" environment (possibly not a VM) as per I would like to have just one server (not PI based systems, that works really fine but.. has some processing limits of course). NB AdGuard as per all required installation base are running fine on 32bit.. so.. yep supervisor don´t really like the 32bit (unsupported version) also if.. well it was working before.. but I don´t rember what for a docker version I had on the ubuntu (unsupported) system...
Run HAOS in a VM - anything else is just going to lead to an unsupported state, based on what you just posted, and at some point HA will break on you
Just updated to core-2022.3.5 and came from core-2021.something Host is a ESXi VM with Debian 10. Now have a unsupported configuration since Debian 11 is requested. The guides say that I should do a fresh install (with Debian 11) and import my configuration, is upgrading the OS not recommended? Why?
Which guides?
Hi @humble mirage, long time no see 🙂 i'm talking about https://www.home-assistant.io/more-info/unsupported/os guide.
That's, as far as I know, mostly aimed at people running something other than Debian.
Hmmm if on my ThinClient I installed Debian, then piHole and HA would that be classed as supported or am I better off just keeping piHole separate?
Unsupported, you should read the ADR/requirements
Ty
If you install anything outside of the ADR then you're unsupported
Fair enough
Allright, creating a backup just in case and storing that locally on my dev machine just in case anything goes wrong. Have a Veeam backup as well but that's several days old now. I'm assuming i need to shutdown HomeAssistant before doing the upgrade? so in the CLI ha core stop? That would stop everything?
That won't stop everything, but there shouldn't be a need to stop HA
Think a 16GB SATA DOM would enough for HAOS or should I just get an SSD? The DOM is already in this TC
I'm currently sitting at 23Gb with a lot of devices and entities logging, if you'd configure the retention for only a few days then i'd say 16gb is enough. I have it configured to 31 days and have 400+ entities.
Hmm k, I have less than 100 entities, and use mariadb for logging. Would be tight I guess
The min spec is 32 GB - though if you're only booting from the DOM and store the data on another disk 16 GB is probably going to be ok
Ah plums, just realised this HP T510 looks like it doesn't support a UEFI BIOS :/
Looks like HA is staying on the Pi for now, and piHole going on the TC. Will move HA to my NUC later
Hey, hopefully anyone can help.
I'm running HassOS on a kvm/qemu vm with bridged network. Now I would like to switch that to NAT instead of bridge. Will that work, if homeassistant isn't on network for real? What nat-rules should I set to make everything working? Any help appreciated.
Depends on what you mean by work really, any integration that uses SSDP, mDNS, and other discovery mechanisms will fail
To update the docker container installation i assume you can just use SSH pull image restart container and done ?
yes that's what i have done so i was correct cheers 🙂
hi, today i hooked up a Monitor to my pi with home Assistant because after an update it isnt accessible and it shows : [WARN] Home Assistant Cli is not Starting. Switching to ermergency Console. Help
what did you update?
the core
a bit unlikely that just a Core update killed everything including the cli
did you just boot it up?
CLI sometimes take a bit longer to come up and you get that warning when it's booting
i left it on overnight to start i think this should be long enough
Sounds tricky.
Maybe you have another idea. I'll use letsencrypt addon for getting ssl cert to remote access my home assistant. But there is another server on the host (where ha vm runs) which listens to port 80. So acme-challenge will fail. But I don't want to use dns-challange and I'm looking for a solution to forward port 80 temporary to the ha vm.
Nat rules with iptables on the host seem not to work because of the bridge interface.
you can use docker ps a to see if homeassistant is running or not or if even multiple containers are down.
then see if ha su logs and/or ha core logs work. If the cli is really down they won't work
if they don't, well we'll see when we got to that 😄
I don't know how to create that for my bot as they as just self hosted open source bots
Why don't you want to use DNS challenge? It works really well
I'm trying to be most independent from external services.
But if theres no other idea, i'll look into it
Well, you're already using DNS for remote access...
Yeah, but all my clients use my own dns-resolver. And thats why I'm trying to be consistent here and not to use a service for dns-challenge.
There is a solution I found. My own DNS hold the dns-zone and can resolve the dns-challange. But I have no own domain. Just a dynDNS name.
Letsencrypt addon does too. I'm using goip and spdyn.
Well, DuckDNS support the required hooks for DNS challenges... if spdyn are any good I'd expect them to as well
you know what? I got why it didin´t work.. CLI should be installed (i386 version as supervisor meaning the script that setup HA is not really bullet proof). Supervisor is "System is not ready with state: setup" for ever.. reason.. docker network NG.. the IP addresses aren´t poperly assigned to the HA supervisor because of Docker (also here the setup part is NG, as per dpkg i should run the docker setup script right?)
Hi, Are we able to restore backups now?
I try a couple of days ago and it worked in my case
It's not intended to be run on something that's 32 bit, or older... Hell, even Docker requires 64 bit these days
You're attempting to push a lake uphill, using only your hands
Hey, i have problem setting up a static ip in RPI364 bit with ha 7.5
I am trying nmcli con mod eth0 ip4.add 192.xxx/24
and i get eth0 error unknown connection "eth0"
I got the right syntax
Hey, I'm running Homeassistant inside a docker container. Now I want to switch from LAN to WIFI. On my FritzBox I have given the Pi a static IP address. The same goes for the Pi itself. Via the GUI I gave the eth0 port a static IP. Is it possible to change?
You can change it back to DHCP and let the router handle it - IMO you should stay with LAN
It seems my hassos installation has been having database issues for a long time. Manually running the recorder.purge service will make the system start consuming 25% of the CPU (VM with 4 cores) until I restart the Core (or the VM). I've got auto_purge disabled for now, but that's not a good solution. I need some troubleshooting suggestions.
yep it is because of Technicians attitude and curiosity.. of course I could simply shut down all and use the PI4 until it will be not working too 🙂 Docker anyhow is working.. only the network seams having problems and I guess it is the OS.. just some more try and then..
well, I'm running a second container with octoprint and my 3d printers location isnt't equipped with LAN. So therefor i want to switch to WIFI. Shouldn't I use the same static IP address for WIFI as for LAN after i deleted the LAN connection. If I'm not addressing a static IP all Mqtt devices will lose the host after a reboot of the rpi.
Update your router with the WLAN MAC address for the DHCP lease and configure the host to use DHCP
sorry took me a while, both are not running ( su and core)
@tardy kelp use ha host reboot then watch for any issues
isnt running either
it seems that almosg all services arent running
any way to start them manually?
You have a problem if this is needed. If the cli isnt running then just reboot
i just reboot now and the first line was no ethernet found is that normal
No
hm but it is plugged in
Unplug both ends, replug, and try swapping for another cable if you have one
ethernet port lights are off? solid?
one is solid the other flashes
Errata Corrige.. After studying a little of Docker networking.. well I can confirm that Docker isn’t the problem.. so.. although it is an official 32bit version.. somehow Supervisor don’t want to start (probably because it is Debian 10 as someone else reported today) .. although.. ordering now a 64bit server and suspended for ever to use the 32bit machine.. from were I can still use the SSD and HDD’s so far.
Thank you. I'll have a look tonight
Hello 🙂 I successfully installed HA Container on Docker (on pi4) by following instructions online. I want to backup my config & update HA but I can't find my config files to back them up. I'm new at Linux but I'm ssh'd into the pi. I just don't know where to look. Can someone help? Thanks!
Did you use Docker Compose files, or a Docker command line?
@sour lark ☝️
docker inspect -f '{{ .Mounts }}' homeassistant
will show the volumes/mounts, you'll see something like
[{bind /data/homeassistant /config rw true rprivate}
Or you'll see nothing, in which case you didn't create a volume, and have a problem
@humble mirage I used a Docker command line. When I use your inspect command, it returns [{bind /clark/.config/hass/config /config true rprivate}]
Then your files are in /clark/.config/hass/config
You should use a compose file going forwards
Ahh, thanks for helping me find those! So, does that mean I can safely update HA and the config is safe and sound?
No, I haven't. Not sure how because the container version doesn't make it easy like the others
what do you recommend?
Do you have other backups for other things?
None of my few containers is being backed up, if that's what you mean
😱
If not, you should look into one or more of:
- rsnapshot, rclone, urbackup, restic, and more
It's all new and I somehow managed to get HA & Node-Red running well, but need some sort of backup system
I'd strongly recommend at least two backup destinations - one local and one cloud. Ideally you want backups that are versioned so that you can recover from a point in time
Yeah that sounds like a great idea. I'm running this on a pi with an SD card (for now). I will install rsnapshot and try to set it up. Possible to backup to thumbdrive attached to pi?
Yup, rsnapshot is fine for that
rclone is great for online backups, and supports many cloud providers. The total data volume for HA is low enough that you can likely back up to B2 for free
Cool, will check it out. Thank you for the guidance.
The only other thing ... remember to test the restore so you know how to do that too
Okay, I have some learning to do but at least I know where the config is now. 😅
When I cd into /.config directory and run ls -a command, there's no such /hass directory. This is where I got stuck last time
Try
cd /clark/.config/hass/config
If that doesn't work then you mapped a folder that doesn't exist, and all bets are off
Ahh, that worked! So I'm not totally screwed?
You tell us 😉
Is that a folder with configuration.yaml, if so, you're good
Mostly anyway
Haha I see all my HA files in there. How come I couldn't see that /hass folder when I was inside the .config?

Yes, configuration.yaml is there
Many possible reasons, from you looking for /hass when you should be looking for hass, to layer 8
Hello all! I've in a bit of confusion at the moment. I've installed Home Assistant Operating System, and it's all working, but I can't seem to get Supervisor to appear?
Look under Configuration. I hear he hides in there most of the time
oh wow. that simple huh? he was hiding in plain sight this entire time. thank you so much haha.
If it makes you feel any better lots of people don't notice it's right there
Thanks. I got it running! Networking wasn't coming up because the SSL config was broken
Nice! 😄
It's amazing when you have working HTTP how much less xxxx is broken! 😋
Is there a neat way I can generate an SSL cert that will be good for mydomain.duckdns.com, homeassistant, homeassistant.local and 192.168.1.123 ?
No
What you can do is use a proxy server for SSL, so that you can use all the other names with plain http
OK. I wondered because my LetsEncrypt cert on my server is for all the domains the web server provides
It has to be able to verify those though, which is tough to do with a LAN IP
AH. Gotcha
Turns out you may not be the only person using those values for a host
Hi, I updated HA but it broke my Zigbee & Z-wave JS contollers. Says it can't connect to localhost. Do I need to do something to remount the USB dongle for those two? I'm on RaspPI4 with Docker
Hi guy's
I'm guessing this is the right place to ask?
How long should it take to update the operating system? 7.4 to 7.5 on 8Gb Rpi4.
Ola all, trying to restore a clean install of hass with no settings. This does not work, any ideas on how to do this?
doing my head in lol... I never thought restore hass with no snapshot would have been an issue LOL
A clean install with no settings is no restore at all.
Semantics away... A clean install within home assistant itself
Don't use the backup service. Install a new image (Hass OS) by following the install instructions again.
If I try to downgrade my HA core version to 2022.2.9, is the supervisor downngrades too?
ha core update --version 2022.2.9
I need to downgrade, because of the broken zigbee ZHA in newer versions
No, just the core. To downgrade the supervisor use ha supervisor update --version x.y.z
It will still auto update though.
but it never worked for me
What does "never worked" mean? What error did you receive?
`[core-ssh ~]$ ha supervisor update --version 2022.02.0
Processing... Done.
Error: No supervisor update available - 2022.03.4`
Yeah it's not actually possible to downgrade the supervisor.
🥳
Ok, downgrade to core-2022.2.3 fixed my zigbee trouble
Anyone know as to why I would be stuck in a waiting pattern for "Preparing home assistant" on a Pi 3B+
Having no internet
It has a constant internet
Yup it is downloading but is stuck on the downloading of the docker image
Hi, Just trying to learn how HassOS works. I upgraded the OS in my RaspberryPi and found two similar partitions post upgrade. hassos-system0 and hassos-system1, and kernel-0 and kernel-1. the 0 and 1 partitions have similar file structures, so I assume they must be the old and the updated versions of the HassOS. Was wondering if I could remove the earlier versions partitions, ie. system0 and kernel0 and still be able to boot.. ?
Bandwidth issues ?
Im on the same setup on 100MBps and setup got done in 20 min
Looks like it was an issue with the image being done in the Raspberry pi imager instead of balena
hey, can I access HA by my local IP if I HTTPS it with cert?
right now I can only access it through HTTPS using the DNS
Yes, you just have to click through the warnings
there are no warnings
This page isn’t working 10.0.0.182 didn’t send any data.
ERR_EMPTY_RESPONSE
I press the i icon and there is no cert or anythign to accept
Did you use https://?
you can add a DNS override on your router and create an A record for your domain that points to 10.0.0.182
then when you're at home you can use the fqdn to get your private IP and not worry about NAT Hairpinning
I did not think of that! That works.
That is very interesting. I do not think I can create any records on my Router. Although, I like the idea
for mine I have an A record for *.mydomain.com so everything mydomain goes to that host
hey i think im getting the same issue as @onyx void and several others. upgraded the home assistant core to the latest. Now stuck at a white screen with stating autoboot. Did some googling and tried pressing 2 and then enter. That appears to go through the normal boot process but then results in returning to the same white screen
Im running HASS OS with a NUC
Upgrades to Core can't impact the OS like that
That'd be like an update to Chrome stopping your laptop from booting
ok, i performed the home assistant core update that just came out. it restarted and now stuck. From what I can tell when I press 2 and enter. the boot sequence appears to be fine with no errors presented.
You don't have to reboot when updating Core
Boot issues are either a problem with an update to the OS, or a hardware issue
Ok, is there a way to dump me to the command shell