2023-01-05 14:57:08.195 ERROR (MainThread) [homeassistant.components.hassio] Missing SUPERVISOR environment variable
2023-01-05 14:57:08.249 ERROR (MainThread) [homeassistant.setup] Setup failed for hassio: Integration failed to initialize.
2023-01-05 14:57:14.515 ERROR (MainThread) [homeassistant.setup] Unable to set up dependencies of raspberry_pi. Setup failed for dependencies: hassio
2023-01-05 14:57:14.516 ERROR (MainThread) [homeassistant.setup] Setup failed for raspberry_pi: (DependencyError(...), 'Could not setup dependencies: hassio')
2023-01-05 15:09:24.552 ERROR (MainThread) [frontend.js.latest.202212131] :0:0 Script error.
#installation-archived
1 messages · Page 15 of 1
Please use code sharing sites to share YAML and errors.
ah i was wondering how you all did that. what site do you recommend?
You've got some kind of integration that's looking for the supervisor or similar. Sounds like a HACS one probably.
i do have HACS installed
The ones in the bot message 😉
oh haha i just ignored that
some of them are humans
The trick is telling them apart!
for images yes, for large logs or code, the sites mentioned
oh i see jsut link to the text like i did?
yeah
i thought i saw people insert lines of code like code tags on ehre
if it is under 15 lines it can be shared here with formatting
if not, then this happens: #installation-archived message
i see
if i go to HACS, id ont see anything ive installed.. i had stuff like terminal at least. maybe the raspberry pi thing was there too
maybe i just uninstall hacs
Terminal and SSH is an #add-ons-archived, not an integration.
HACS only does integrations 😉
ohh
if i open HACS would it be clear if i had anythign installed? I dont see anything. jsut a list of repos
oh wait HACS i think is how i got Adaptive Lighting. so thats why i have it
so the integratoins indeed just show up in integrations
that fixed it
welp, this migration has been a success. what a day.
thanks for all the help
hello, does any1 have a working installation tut for 32bit intel processors?
The documentation does 🙂
topic, top of your screen
E.g. https://www.home-assistant.io/installation/raspberrypi is the same, just download the 32-bit versions instead of 64 😉
Thank you 🙏🏻
I'm pretty sure there's no HAOS support for 32 bit Intel any more
I'm not sure there's even Docker support
Oh, no, there is
Of course, there needs to be Docker support for i386... and that died ages ago
yeah, found that Docker stopped supporting i386, kinda dead end from there
You can pip install (aka Core) on 32 bit
If it's an Intel 32 bit processor, it must be ancient
This is not a search engine, if you have a question you might want to ask it 😉
@odd marlin testdisk is able to find hassos-overlay and hassos-data, and I can even browse through the directories. I have no idea what to do now, though.
Oh, and it also found three partitions all called rootfs, but it said they can't be recovered. Idk if that's important or not.
Those partitions are probably false positives since the data disk tool just creates a single partition. Try doing a find /path/to/mountpoint -iname "[a-f0-9]*.tar" and see if you find any backups
What I meant is, inside testdisk I can select the partition and browse files that way. I still can't actually mount it regularly
oh I see. Is there a folder called supervisor?
supervisor/backup
Nope
apparmor
apparmor, manifest.json, repositories, and a whole bunch of folders with random hex strings
The hex strings may be folders it couldn't recover the name for
Can you copy all the data out of testdisk so you can search through it with find or similar?
I'll try
There's 2 tar files in /supervisor which seem to be corrupt
But the random folders are all empty
They're both too small to be backups anyway (<2kb)
😦 Kinda running out of other options if all those recovery programs weren't able to find real backup files. It might be time to go with fresh install, as painful as that might be
🧌
Step 1 after install: Set up a backup system that offloads the backups elsewhere
You definitely didn't have it uploading backups to say Dropbox, google drive, an SMB share or something?
I was planning on setting it up with smb, but then I got frustrated with user permissions and stuff and then never thought of it again 😭
The last manual backup I have is of March last year, but I've changed quite a bit since then
Oh well...
March of last year is better than starting from square one at least
True
At minimum you can use it as a reference for a starting point
In addition to enabling offsite backups, I'd recommend putting the Pi's data drive on an external SSD in the future rather than a USB flash drive
Glad to help. It's unfortunate that the drive managed to get that badly mangled 😦
virtualbox -> debian linux -> haos_ova-9.4
I run it. sometimes i get "failed to create a file system, ext 4"
Most of the times i just get a black screen.
I have tested
increase videomomery
v-sync enabled/disabled.
Other things i can do?
Debian > HAOS
How does that work 🤔 HAOS is it's own os
Did you use https://www.home-assistant.io/installation/linux ?
It has a ready to use vdi for VirtualBox
yes
new error. i can choose from
Slot a ok=0 try=0
slot b ok=0 try=0
slot a rescue shell
slot b rescue shell
All shows
error: no such partition.
press any key to continue
Are you sure you enabled EFI?
i have spent another hour searching for a solution but i can't seem to find how to get a bootable partition. i can enter grub menu but that doesn't give me much (nothing)
Any tips on fixing error "error: no such partiotion" ???????
Can anyone help me out with this: everything i try to enter it just says reload. My whole HA is broken after installing a backup. 502 Bad Gateway
nginx. How can i fix this?
hey Hachiko -- i had the same problem. Any restart from HA doesnt seem to work. If you unplug your power from your device and replug it in, basically power cycle it, it should work again just fine
thanks for the advice but unfortunatly it did not do the trick
i've been googling/searching reddit/community for a few days but can't quite get the answer
i've got a
-RPI compute module 4
-180GB SSD card
-8 TB HDD
i've got Usb boot working great, and i've got HA OS on the SSD (and all of my add-ons, integrations, automations working great!)
But any time i plug in my HDD, nothing works. Seems like maybe the RPI is trying to boot from HDD? Is there a way to say
"Please boot from SSD, don't boot from HDD" ? If there is, i can't find a way. I'm 'simply' trying to use the RPI to run HA, and then have it also do Plex
First of all, what's your power supply like?
2 separate AC adapters
- HDD outputs 12v
Because if you have too many things plugged in that draw power, your Pi is going to be cranky
- RPI outputs 12v
Powered HDD is good!
Alright, so you need to set the boot order for devices in the case of multiple USB drives, I seem to remember that involves using an install of Raspbian on a microSD card, and rpi-eeprom-config, but I don't remember more I'm afraid
okay! i can chase that down! thanks for the suggestion
Can you access HA directly using the port and 8123?
i found in my log this msg, how can i fix it: ERROR (MainThread) [homeassistant.config_entries] Error occurred loading configuration flow for integration apple_tv: /home/thoht/.homeassistant/venv_3.9.2/lib/python3.9/site-packages/_miniaudio.abi3.so: undefined symbol: _atomic_load
curiously i don't have any apple tv
Hello. I fail to restore a backup for the purpose of migrating from VM to physical RaspberryPI. The full backup is of size 40MB, it states that 3-4 folders and 2-3 addons are included - however - after restoring the backup the brand new install just keeps looking like a brand new install and nothing of what is showing on the original VM install shows on the Raspberry install. I've googled a lot and have found others with the same problem just no solution. How should I move forward, what can I do to make it work or isolate the issue so that a fix can be filed. Thank you.
the appletv integration picks up pretty much everything which provides airplay features, ranging from tvs to speakers and also smartphones can act as airplay receivers
what installation do you run? haos or else?
I'm running HAOS on the VM, downloaded from the HA website as an .vmdk and HAOS on the RPI following the install docs from the ha.io website.
hm, that should be pretty straightforward then
Hi, i have a very strange issue. Probably somebody has an idea? https://community.home-assistant.io/t/after-a-certain-time-supervisor-communication-from-core-seems-to-stop-working/513971
Nah, had duckdns & nginx - single url
Yes. So you should still be able to access it via the IP address if youre using a reverse proxy
Still cant access. Err_emtpy_response
Are you using [local ip address]:8123
Yes.
What install method do you use?
VM - proxmox.
depends on what you mean by "right".
ipv4 is right.
Right as in the IP that should be set static DHCP in the router & what you expect
then it comes up docker ip adress, dns and gateway. That is not the same as in router, but the ipv4 is the same as in the router..
Yeah ipv4 is what you're wanting
You might need to check the logs on the host itself
Thank you for the advice, but how do i do that ?
hi I'm trying to install UniFi in my Ubuntu VM, I finished everything but it says 'Authentication is required to restart 'unifi.service'
how do I give this service authentication? thanks!
is there a way to give static ip to the vm in proxmox? if so how do I do that?
Just add it as a static DHCP lease on whatever is doing your DHCP stuff
Okay, thanks
Hi, I live in multi story home, and I have one BTHome sensor, and I get multiple times a day that new sensor has been detected and they are just random MAC adreses
Is the a option to tur off auto discovery for this one integration ?
on my BT sensor I have the System Options for bluetooth (under the three dots of the general BT integration)
there I can enable / disable automaticallyo adding new entities
not sure if this is the same bt integration as you are using though
but turning it off might be a good idea, since I now have the BT toothbrush of the neighbours on my system apperently
Thank you, will try to do this, for now i tried to disable this same setting on BTHome integration, but that didn't help
Updated python 3.9 to python 10.0, set pip, pip3 python and python3 commands to use python 3.10 instead of python 3.9, updated home assistant and i get these errors:
Jan 06 12:55:29 rpi4-20210823 hass[529365]: 2023-01-06 12:55:29.208 ERROR (MainThread) [homeassistant.setup] Setup failed for analytics: Unable to import component: No module named '_sqlite3'
Jan 06 12:55:29 rpi4-20210823 hass[529365]: 2023-01-06 12:55:29.220 ERROR (MainThread) [homeassistant.setup] Unable to set up dependencies of onboarding. Setup failed for dependencies: analytics
Jan 06 12:55:29 rpi4-20210823 hass[529365]: 2023-01-06 12:55:29.221 ERROR (MainThread) [homeassistant.setup] Setup failed for onboarding: (DependencyError(...), 'Could not setup dependencies: analytics')
Jan 06 12:55:29 rpi4-20210823 hass[529365]: 2023-01-06 12:55:29.223 ERROR (MainThread) [homeassistant.setup] Unable to set up dependencies of frontend. Setup failed for dependencies: onboarding
Jan 06 12:55:29 rpi4-20210823 hass[529365]: 2023-01-06 12:55:29.224 ERROR (MainThread) [homeassistant.setup] Setup failed for frontend: (DependencyError(...), 'Could not setup dependencies: onboarding')
Jan 06 12:55:29 rpi4-20210823 hass[529365]: 2023-01-06 12:55:29.231 ERROR (MainThread) [homeassistant.setup] Setup failed for recorder: Unable to import component: No module named '_sqlite3'
sqlite3 is installed, but the python module seems to be missing
and i bet after I fix the sqlite3 module, more modules will be missing. so, my question is, which step of the python upgrade did i miss?
The part where you have to create a whole new venv?
i did
You deleted the old one and created a new one?
i deleted the whole folder of venv
yes
i basically did a brand new HA install
of the venv
I wonder if your Python wasn't built properly then
Did you build Python from source yourself?
ok i did this and now i get 404 sudo apt-get install libsqlite3-dev
What OS?
i used this guide. and then i used the alternative installations command to change my python commands to use python 3.10. pip, pip3 also use python3.10
That isn't necessarily going to give you all the required packages for HA
WHAT OS?
debian
Debian WHAT?
bullseye
i did that before i updated ha but let me try again then
$ apt search libsqlite3-dev
Sorting... Done
Full Text Search... Done
libsqlite3-dev/stable 3.34.1-3 amd64
SQLite 3 development files
no new updates
libsqlite3-dev/stable,now 3.34.1-3 arm64 [installed]
SQLite 3 development files
Right ... well... it's installed
Maybe this is a sign that you should use Docker
And by maybe I mean it is
Trivial
ok so now instead of the HA port being dead, it shows 404. so something has changed
let me check systemctl
ALright, let me try that
installing docker
do i create the /docker in the root of the file system or in the .homeassistant folder?
/docker is explicitly in the root, hence the starting /
ok thx. just asking to be sure
Creating it in .homeassistant/ would be ... unwise
Also, do remember to add /docker to your backups
Or set up backups
Then test them 😉
Hi everyone, i'm trying to set up my Glances to be automaticlly running whenever VM is back online, but i'm keep getting error in the last stage of this guide
https://www.how2shout.com/linux/install-and-use-glances-to-monitor-ubuntu-20-04-server-or-desktop/
I get connection refused error when doing docker compose. wth
@sharp cipher I converted your message into a file since it's above 15 lines :+1:
restarted docker service. lets see
used docker compose up from the manual. hopefully its not something completely different
seems like up pulls the containers and then starts them
i did sudo su homeassistant so only step i missed was to stop the ha service. which i will do now
@humble miragechange the command
docker-compose logs --tail -100 -f homeassistant
to
docker-compose logs --tail 100 -f homeassistant
It successfully started! Thank you!!!!
will it start automatically on reboot?
Try it
i saw one of your commands had -d flag. which is for daemon i suppose?
Tinkerer mind helping me with Glances? I can't get it working automatically after reboot VM
It'll restart on reboot if you set everything up correctly, but the only way to know you did is to reboot
@woven turtle Please do not cross post. Read the channel description, post it and wait for folks to respond. Crossposting wastes people's time as they're unaware of the help you're getting elsewhere.
If you don't get any responses after an hour or more, and your message is no longer on screen, it is fine to re-post or post a link to it.
let me reboot then
I saw your post in the relevant channel... if you'd been patient I may have even got around to helping you...
Thanks!
yep! it is working!
Thank you so much
seems like the HA boot is faster than before
Now my ultimate question is. Is it possible to bind the update button in HA system menu to run a shell script that does docker compose command?
So that i can update HA from within the app and not having to ssh into the Linux machine
hi, in your opinion, is it better to include my garage and my house into the same HASS instance or create two HASS instances linked with remote assistance ? (I have 2 distinct power entrance, lots of devices in house and some in the garage)
If it's the same property, then one instance tbh
Hey there. I did a fresh install of HA on Debian 11. All went well. I am trying to configure static IP address on the server. I tried putting static IP in "/etc/network/interfaces", however this file appears to be ignored. Where do I setup my static IP address in Debian?
You're better off just setting static on the DHCP server
So there is no way to install home assistant directly on macOS anymore?
@manic bane That complicated huh?
No just best practice nowadays
I see. Well, is /etc/network/interfaces not where I would set it up if I wasn't going the DHCP route?
sure, my script is a simple docker-compose pull && docker-compose up -d
Thanks, I will then merge my 2 power entrance into my energy panel ?
So there is no way to install home
But how do u run the script? By ssh-ing into the machine and running it whenever a new update is live? Or have you somehow bound it to some button or service in HA?
Surely the update button calls a service which only exists in HAOS or supervised installations, right? So perhaps by creating that service in a config and making it execute those commands in a shell would yield the same results?
Idk if it's possible to view if clicking a button in HA calls a service or not
I have home assistant on proxmox and proxmox says its using 6gb ram while home assistant says it uses 6% of the 8gb ram, how can i fix this?
You can't
It's Linux using cache/buffer
The OS frees the cache/buffer when programs need it
but is 6gb ram not to high
So technically the ram isn't reserved. It's just used to make it more efficient
No
Then what is it that makes the diffrence so big between the home assistant dashboard and proxmox
proxmox 42%, ha 19%
HA shows the ram used by programs etc. Promox shows ram usage including cache and buffers of the OS
It's really not an issue
If a certain program needs free ram, the OS takes care of it and will flush something from the cache to make space
It's to make the system run smoother
oki, Thank you so mutch!
By caching frequently accessed files, IO etc
Also, are 2 cores enough to make home assistant run. Im running on my nuc also freenas, or should i both max then at 4 cores. since ha not always using much cpu
2 cores are fine. That's all I've got allocated to mine as well.
type: conditional
conditions:
- entity: binary_sensor.docker_hub_update_available
state: 'on'
card:
type: button
tap_action:
action: call-service
service: switch.turn_on
target:
entity_id: switch.updateha
thats the whole card, I use a switch because I can provide it with a command_timeout
Hi!! I'm new in this chat and new with home assistant and pleaseee I need some help to install it and a lot of patience because my knowledge is limited 😅 I have created a virtual machine on my Synology NAS and installed Home Assistant. So far so good. But I can't access the dashboard http://homeassistant:8123/
any idea??
Try the IP instead
homeassistant - or more correctly homeassistant.local - requires mDNS working
It does not work either
No, the IP and that port
I'm going to try
The IP of the VM
http://homeassistant.local:8123/
Cannot access this website
Check homeassistant.local for a write error.
DNS_PROBE_FINISHED_NXDOMAIN
what browser u uisng
I don't know if its a problem with my router. Maybe??
homeassistant.local requires fully working mDNS
Your VM is using bridged networking?
mmm... I dont know what is a bridged newworking 😅 @humble mirage
it's a VM created in a Synology NAS
Well... I know nothing about Synology...
The good news is that you can connect to the console of the VM and see what's happening
I'm trying to do it. It says "booting from had disk..."
It should say more than that
Then it's not even booting, which would be the problem
Make sure that the VM is configured with UEFI support
I'm going to check it
Legacy BIOS
i need to change it, right?
great, thanks, Im going to do it
@humble mirage Now it works!!!!! Thanks so much. Maybe it's silly for you, but it has been a great help. Thanks so much ❤️
To be fair, the alternative install docs aren't that clear on the need for UEFI - it's there, but it's far from obvious
I followed the documentation, I don't know if it was clear or not; maybe I was wrong. In any case, it has been easy. Millions of thanks
Since it's this easy, is there a reason it isn't natively supported by HA?
If you want that, use HAOS
Then you too can mash the upgrade button without reading the release notes, and then wonder why HA is broken
Do you actually get list of updates in the settings when using docker? I didn't get that with core
@low notch what's the code for docker update available binary sensor?
Just using rest with some docker repo to check for latest ha release?
No need to view your configuration. Plain words is enough of an explanation
Version integration
There is no way for a implementation for every situation the containered process runs in
There's just 4. 2 of which do get automatic updates.
Leaving you with docker or venv
But then there's the issue that HA can't update stuff like python for you and recreate venv
Though docker does that right?
Or is there any other reason why it's a bad idea to allow automatic updates for docker install method?
Thank you. Didn't know that was a thing. Though i did have a preinstalled card like a year ago that showed whether or not an update is available
Automatic updates means that at some point you'll wake up to a dead/broken HA
I meant semi automatic. Like being able to risk it and do it through the UI. I had 0 issues updating for the whole of 2022. Only issue was with 2023 due to new python version
Though what can break with docker? Since the pulled image sets everything up automatically?
Yes
Blindly updating any install runs the risk of things not working
HAOS (and Supervised) are the only ones with a built in UI to update with
Well, i won't be blindly updating. DW.
The release notes are (mostly) for core, which is common
Which is not be be confused with Core
Yeah. And that's why I'm asking why docker doesn't? Since the docker image does basically everything for you
So the only thing that can go wrong is breaking changes with integrations. Which also affects HAOS and supervised
The Docker image doesn't do basically everything for you
@low notch I've got a question... You clicking the button, which calls the service to turn on the switch... Wouldn't that switch stay on even after reboot? Essentially making an infinite loop? Or does that not fire on initialization and reboot isn't counted as state change?
Perhaps you turn the switch immediately off using an automation?
I can't access my homeassistant anymore. It's been a while the rpi has been unplugged so Im considering reinstalling the firmwarew. So I have two questions: Do you know if there's an easy fix for this? If not, what file should I keep for the reinstall ?
Backup the .homeassistant folder
Hmm I guess I need linux to access it right?
Because I don't see it in my usb drive on windows
It's on the same drive as your Linux OS
You were booting your RPI from a USB drive?
Anyways no. If you have w11, you can use use wsl to mount ext3/etx4 drives
Nono, I removed the SD card and plugged it into my pc. I couldn't even connect via ssh to my home assistant
what do you mean ?
That switch defaults to status false, it is only on in the second i trigger the button
Ok good
Im on my Linux system now and I can't see any .homeassistant
cd /home/homeassistant/.homeassistant
Does it exist?
If so, then do cp /home/homeassistant/.homeassistant /whateverpath -rf
Im kind of worried rn haha
I have 4 drive : 25MB Volume
268MB Volume
Hassos-data
Hassos-overlay
None have a folder home with something in it
Im not on the rpi itself, im on archlinux of my computer
With the sd card plugged in
not sure, what happens if it boots?
did you connect a display and see what it displays?
But I'd rather just reinstall hassos
Hmmm no I didn't. My computer is upstairs, my ethernet cable are downstairs
those would contain everything including the addons
Hmm I don't have anything... I was trying to set some things up with hassos, and then gave up. I wanted to pick it up today, but the rpi just won't connect anymore
I had set it up for wireless connection tho. Maybe something went wrong with that ?
most likely its super unreliable
Do you know how I could remote the setup and make it work with ethernet?
I thought simple plugging the ethernet cable would be it but it didn't.
Otherwise, I'll just start fresh.
I didn't have much in the first place
you can grab the stuff from the supervisor/homeassistant directory, thats the homeassistant config directory
it contains a bunch of hidden folders and files though, so make sure to copy those too if you want to put that stuff back
Thanks, I'll try to do this
Yikes I can't move them
Nor copy. It says I do not have the permissions for it
I'll just fresh reinstall it. It will be easier
Im trying to upgrade core and it not going well
What's the reason that you're running core?
I ran this command to image the SD card. Put it in and booted it up. I don't see anything popping up on the network. sudo dd bs=4M if=haos_rpi4-64-9.4.img.xz of=/dev/mmcblk0
So I don't really want to meddle around with this because I'm sure the maintainers of HAOS know what they're doing, but would you advise against turning off swap?
I believe my entire host just locked up because I ran out of swap
This is actually a common problem I run into on Linux and I think I've had success in the past just disabling in
This is my free -h output from a ssh onto the host (not ssh addon):
total used free shared buff/cache available
Mem: 3.8Gi 3.7Gi 34Mi 3.0Mi 85Mi 68Mi
Swap: 978Mi 978Mi 0B
I'd advise allocating more RAM or figuring out why you're using so much if you're not expecting it
Disabling swap wouldn't help with what you described. That's not a healthy memory usage situation
Can't allocate more ram, and yes I would like to find the root cause. I wasn't able to run the profiler though since the machine was at 180+ load avg
Wouldn't it prevent a lock up?
My CPU usage was pretty low. I'm pretty certain it was locked up due to disk usage. No disk usage if it isn't swapping from swap
Yes I agree, but I'm also not understanding why you think my computer would still lock up with swap turned off?
Because I believe the lock up was due to swap being hit constantly
Your system likely used swap because it was running out of memory
You're focusing on the wrong thing. Figure out why you're seemingly short on RAM first
Well if it happens again I'm not going to be able to troubleshoot with a system that takes a minute to register a keystroke
If you're swapping, you don't have enough RAM. The fix is to add RAM
Uh I mean for all I know it would have kept using up twice the ram if I could double it
The fix is to find out what ran away with all the ram like @brisk willow mentioned. But I'm not seeing justification for leaving swap on
What do I get from swap? Another gig of RAM. So how about I add another gig of ram and turn it off?
Possibly, or it will use the additional memory for buffers/cache which is a good thing
But I'm pretty sure 4 gigs is enough
It's clearly not unless something is acting up
Have you made any changes to your system lately?
Nothing I can think of
I'm going to add more visible metrics for my memory usage though
Though I restarted somewhat recently so I don't think it was something that happened gradually
Actually my system is back up. Let me take a look
Does anyone know if the memory_use_percent sensor from systemmonitor includes buffer/cache?
Actually it does look like it was growing slowly
I restarted HA multiple times so I'm thinking it's an addon then
it does not, because buffers/cache are considered available memory
Yeah ok. Just making sure
But still going back to the swap discussion, I'm not understanding why I would want to leave it on?
I'm gonna add that sensor now actually, there's a swap one
because if you have insufficient RAM, it could just lead to more processes killed because of running out of RAM
it really just leads back to figuring out what's using RAM inappropriately
Yeah but if my system locks up I have to kill every process when it reboots. I'd rather have processes killed with my system still up
i have swap disabled on my server, but also because i have 24 GB of RAM and only 4 GB are actually used; the rest is cache
Either way I'm going to have issues, but one results in a still usable system
And I have more than enough memory than I need
You clearly don't if you're swapping though
okay, go ahead then
No one would. It would just be a matter of time
If you know you will never run out of memory, then it's fine to not have a swap file/partition. If you turn off swap and end up running out of memory, instead of getting slow you'll just have things start crashing
that's not even close to being true
I dunno, I've never been on a system where it was more of a problem that procs were dying cuz I was low on ram vs all the times my whole system locked up cuz of swap and I had to restart
'Locked up cuz of swap'??
Looking at my usage the baseline is about 50%. So yeah I do have more than enough RAM. If you're saying it's because of this issue, more ram is only delaying me finding out I have a memory leak
Essentially locked. load avg was 180+
If for some reason my only two choices were my system becoming incredibly slow and nearly unresponsive, or having important processes killed by oomkiller, I'd pick the former
But in both situations the processes are being killed because in the former case you need to restart
And if you're worried about me running out of memory, what would you say to me adding another GB of ram and remove my 1GB of swap?
That would fulfill your reasoning right? Yet that doesn't seem right to me
this really all just goes back to figuring out why you're running out of RAM
It does, but I don't like my system locking up like this where I can't troubleshoot and need to restart
Load average on the CPU?
Looks like supervisor supports tracking addon cpu/mem usage but it's disabled by default
I don't think it was the CPU since usage was low, my swap was maxed (main indicator) and there was a kernel thread which I think is related to disk activity had the high CPU usage (kblockd)
Is there a way to turn on all the addon usage sensors at once? Or have them enabled by default?
What are the specs you're running HA on + what else are you running other than HA
I'm on a NUC running HAOS in a VM with 4GB RAM
Nothing else
Unless you mean like addons? Not sure what you meant by HA. Like HAOS or core?
In any case, I don't think there's anything else to troubleshoot atm. I'm enabling addon resource usage sensors and I'm gonna watch mem usage
if you're running haos, look at output of "free" and "docker stats" in the cli
Yes add ons
Also running any other VMs?
Nope
So no addons or any other VMs?
Sorry, yeah I do have addons
Quite a few. I was just enabling cpu and mem sensors on them and I added them all to a grafana dashboard
Yeah docker stats wouldn't even load. But if you read further up I did look at free and that's what made me realize it was the swap
tldr. but why wouldn't docker stats work? are you using non-privileged ssh plugin?
I mean the actual console of haos
See what's leaking
eg mine is https://imgur.com/2FBf2JS
They didn't work cuz my system was just too overloaded. 180+ load avg
took me a few mins to run free lol
sounds like something is fundamentally borked. Disable all your addons (especially hass ones) until you system returns to sanity.
Which ones
If you look at the pic I posted above, mem was growing over a long period
Just gonna monitor addon usage and I think that'll find me the culprit
Off to bed, thanks for help everyone
Hi there. Since a number of releases of HAOS, the Login button does no longer react to the enter key. Was that removed intentionally?
Keyboard shortcuts are a mess, just try navigating with enter, tab and escape in modal dialogues. Hitting enter in a input just closes the modal while escape does nothing (most of the time, sometimes it does close them). So there are huge inconsistencies in this area
I totally agree with that one. Huge mess. But when logging in I think it was a good one. 🙂
@low notchwhat have you done? made an sh script inside your /docker folder and run that sh script or running the docker-compose command directly from HA? as i get error code 127 (command not found)
my pi 4 wont show up after flashing it any fix ?
also, dont sh scripts have like an execution time limit?
because my internet be slow 😄
any idea how to completely remove unifi from my VM (Proxmox)?
been trying
sudo dpkg -P unifi
but i still see some unifi files
it also sais i dont have a internet conection but i have a lan cable inserted
Did you install unifi directly on Proxmox? Or onto a VM
I did a switch, with a command_on and command_timeout, nothing spectacular, no scripts
into VM
Then what OS is the VM on? And can you just nuke the VM and restart
Ubuntu
can't nuke it becasue I got some other services installed on this specific VM
Thanks, but I wrote that I have done this command and the files are still there..
Did you restart?
I did.
Then remove the files I guess
This isn't a unifi support discord. Youre better off asking in a more specialised area
by the way, do you recommend me to install all those services in a Docker instead straight to the VM?
Since I am guessing this isn't an add on you're trying to remove
Yes docker makes things easy. And removal is easy
Or LXC
I do
sent you in DM
has anyone here got HAOS on Truenas scale as a vm, I followed this guide to no avail https://www.jcreek.co.uk/home-server/truenas-scale/installing-homeassistant-on-truenas-scale/
What issues are you encountering
i get to the end. fire up the VM and nothings on port 8123
What does the VM console show
just shell
What specifically, does it show the IP address ?
With an ha > prompt?
Also... we're on rev 9.4
ah i think i figured it
dd says no space left on device
how big is HAOS now? i allowed 50gb
Is that possibly split up where it wouldn't be available
not at all haha nothings on it
its a 400gb ssd i just slapped in
would you mind if i DM and go over this in steps
Feel free to continue in here, its what it's for 🙂
no problem just didnt wanna spam it up with screenshots and such haha
You can't post images directly, just links
ok starting from scratch now new ZVOL
wget haos_ova-9.4.ova
done
so next i qemu-img convert it right
to hassos.img?
so id take home-assistant.vmdk and qemu-img convert -f vmdk -O raw
"root@stratum[/mnt/IOCAGE/HomeAssistant]# dd if=hassos.img of=/dev/IOCAGE/HAOS
dd: writing to '/dev/IOCAGE/HAOS': No space left on device
33+0 records in
32+0 records out
16384 bytes (16 kB, 16 KiB) copied, 0.176787 s, 92.7 kB/s"
16 kilobytes
somethings wrong then because truenas says its empty
oh i see
size for this zbol is 16kb like jumbers says
so set force size 50gb?
50 GiB
yay its working now hahaha
noob mistakes
is this correct?
looks reasonable
theres all the settings, i have got this correct right
only a 1mb video settings on the link
its booting i can see it in the display
works! thank u @fallen lichen
I used this download to get homeassist. https://github.com/home-assistant/operating-system/releases/download/9.4/haos_rpi4-64-9.4.img.xz . I then flashed it with this command. sudo dd bs=4M if=haos_rpi4-64-9.4.img.xz of=/dev/mmcblk0 I put the memory card in and plugged it up. I can't see any signs of booting. I don't even see a bios boot up coming out of the hdmi 0 port. I have both the green and red light solidly on. I am not sure what is wrong.
You are trying to boot on a Pi4?
Yes, pi 4 model b
if=haos_rpi4-64-9.4.img.xz
``` 
Yeah... you flashed a compressed image
That's really not going to boot
xarchive cannot uncompress it.
Hi. Hope someone can assist, I've been looking for help everywhere and can't seem to find anything on this... I've just installed HA in a linux VM; everything worked fine, no errors booting up the server. But it did not detect any of my smart home devices automatically, and adding devices manually does not work, either. It does not seem to be able to reach them. For instance, for Philips hue it says "Unable to discover Hue bridges". Any idea, please?
You need the xz tools
Your VM... it did use bridged networking?
yes, bridged adapter, Promiscuous mode set to Allow All.
I have noticed that if you give HA only a reverse proxy connection, it can't find devices. It needs a insecure connection to the rest of your house. Basically, a bridge to the network.
@daring lance Sorry but I'm not sure I understand what I should do
How is your network connected to the HA device? It is connected directly to to the network or what?
I'm running the VM on a mac wired to the router
the virtual machine needs to be given access to the network with all your devices. Look at how to connect a virtual machine to the network. I assume that virtual machine has no internet connection at all.
I found the issue. I have 2 bridged routers, it will not work on both of them, just when wired to a certain one. Thanks for trying to help
I'm running HAOS on HP t510. HomeAssistant logs (recorder) in a PostgreSQL database running on another system.
Yesterday there was a power failure and when the power was restored HA started, but the database server wasn't ready yet. So HA complained loudly.
My question: does HA recover recording gracefully when the database is reachable again?
hey there,
I set up nginx to forward https request to http request in HA.
Normal reverse proxy config.
I am now able to access my HA instance from internet but after inserting credentials I fails in login
Message from log is "HTTP Integration is not set up for reverse proxy", where do I do this and how do I do it properly?
☝️
In the docs for the http integration
I thought so, but did not find the integration within HA Web UI
But I will study the docu now. Thanks
Not all integrations can be configured in the UI
Yes, but I only knew from them which were displayed in UI. Did not know there are others which are not listed
The UI list does tell you that there's more ...
Oh, interesting... they removed that
That was ... short sighted
|| Are you surprised? ||
Some in the intergrations like like CalDAV tell you they can't be done from the UI, others like HTTP just aren't there
¯_(ツ)_/¯
Maybe because HTTP is considered a system configuration, rather than a true intergration? "Logger" also isn't there
only the path to the ssl certificates is not recognized properly.
Invalid config for [http]: not a file for dictionary value @ data['http']['ssl_certificate']. Got '/etc/letsencrypt/live/my.dom.ain/fullchain.pem'
not a file for dictionary value @ data['http']['ssl_key']. Got '/etc/letsencrypt/live/my.dom.ain/privkey.pem'. (See /config/configuration.yaml, line 12).
Are you running a Core install?
Or are you running Container (Docker), HAOS, or even Supervised?
SuperVised

Installed directly on linux (arch)
Well, you'll not find the certificates in that location inside the container
Well, you win today's most broken Supervised install award
well needed it to get homematic running
No... you didn't, you had other options
Well, what I found was, that I need supervised to install addons
HAOS -> got no system for that am not willed to get another one
Copy the SSL certificates somewhere under the config folder and use that path
is a symbolic link fine as well?
No
Because a link says the file is here and that location is outside the container
aalright
You should just switch to a Container install before it all blows up
I was never able to successfully create a virtual machine and get a guest running, under linux. That was why I decided against it.
Well, be prepared for HA to eventually just stop working
would be not a problem right now. Even switching to a container would be possible (from what I did in HA till now). So I think I will give it a try then.
Do you recommend docker over VM?
No, both Container or HAOS in a VM are fine. Whatever you prefer
had trouble with both of them. So it doesn't make a difference 😄
Switching from Supervised to HAOS takes probably < 30 minutes (as long as your Supervised install hasn't yet blown up)
Yeah maybe. But installing docker and getting it running will be the troubling part (at least for me).
Porting to HAOS is not necessary. I can simply do everything from the beginning (I only installed Homematic so far, no dashboard configuration etc.)
https://www.home-assistant.io/installation/linux#install-home-assistant-container
did so and it is running
hi there. I've just installed the Watchman integration and it reports to me many missing entities in the file lovelace.lovelace_home which is stored in .storage. I'd like to know what this folder is and what it is used for ? I'm running HASS.io.
I'm trying to install HA Container on Pop!_OS, but when booting it seems to be stuck on ```
INFO (MainThread) [homeassistant.core] Starting Home Assistant
My configuration is pretty straight-forward:
services:
homeassistant:
image: homeassistant/home-assistant:latest
volumes:
- ./config:/config
- /etc/localtime:/etc/localtime:ro
environment:
TZ: Europe/Amsterdam
restart: unless-stopped
privileged: true
network_mode: host
Has anyone managed to get Home Assistant OS installed as a VM on the latest TrueNAS Scale?
How does one install esphome firmware? I have a 4-pack of Gosund WB4’s. I’m having a friend solder the serial wires and flash the firmware.
Wondering what program to use to flash?
this may work for you. haven't messed with it in a long time https://github.com/esphome/esphome-flasher
@rough falcon I've gone through that but I'm running into a screen like in this post https://www.truenas.com/community/threads/vm-add-existing-disk.98140/
im stuck at the boot screen
my screen is blank when i try boot from file
to be clear, you mean when you click display from TrueNAS ui
yes when viewing from the webui using either VNC or Spice
gonna try something real quick
ok its working now. i did something stupid. kinda confused myself and copied the image wrong. was working off of a few different guides to get the usb working as well
thanks for the assist. made me reread everything
nice, glad its working now
trying to migrate away from using proxmox
shoulda just stuck with your walk through
Thanks, I try to keep up with things to make sure it works
well thanks to you I don't have a use for my proxmox setup anymore. so one less space heater running
Hello. I want to buy a nest 2nd gen wired camera. Have I understood correctly that this camera CAN stream via webrtc directly in HA but 1) i have no media source when it triggers an event (like no screenshot) and 2) i cannot record via HA locally if there is a trigger?
I am probably wrong...otherwise why to buy this camera?
Reposting from general; suddenly lost access to my HA; I thought it was a cert issue but I put updated certs there and restarted and I still can't access HA. Any thoughts on how to troubleshoot? I can SSH in to it.
What kind of "can't access"? Can't open web page, can't login?
can't open page at all
Unable to connect to Home Assistant. Retrying in ## seconds
What about the console directly?
I can SSH in.
core logs and supervisor logs show that it's doing its thing, is there any way to view access logs?
Got it, blew away the certs again and copied them over again and rebooted the machine.
It seems like HA is extremely serious about having valid certs, I can't even access it with a warning if something is wrong with SSL
SOmething is probably failing to load, which is breaking all the things.
Wish I knew how to track it down. I can reproduce by just removing the certs from /ssl and then the whole thing just dies
Can't even access it locally
nothing in home-assistant.log?
Hello all. I use HA docker (with docker compose) and would like to restart another docker in my automation. I have mounted docker.sock already but my issue is that my HA docker doesn’t have Docker installed (« unknown command docker »)
I’ve been googling a while and find no easy way to enable docker command in my HA docker. Any idea ?
All I found is pointing toward docker-in-docker (dind) but I’ve been unsucessful to add it to my HA’s docker compose so far
I’m guessing it’s a known topic for one of you with a simple solution
make a shell command that runs a command on the host via SSH
Ok i tried to use shell command from within the container 😭
I look at your link. As usual i spent 3hrs on google and when I give up you give me a 2 line solution
You are the best guys
https://blog.ceard.tech/2020/05/home-assistant-docker-and-using-ssh
ssh -i /config/ssh/id_ed25519 -o StrictHostKeyChecking=accept-new tinkerer@orbital.ceard.tech docker restart homeassistant
(clearly, don't use that actual example command... even if you can reach my orbital mind control lasers, you don't want to restart HA that way)
Lol thanks !
Actually its to restart an add-on (ran as external docker) which has a bug… so i need to restart it whenever HA is restarted. Until bug is fixed 🥶
Startup automation to restart HA the add-on, what could do wrong...
It’s not really listed as an add-on because i have docker setup
So it’s just an external docker

I mean i installed an add-on manually because its not supported so it became an external docker
I think i have no other alternative
Its a standalone docker pushing data to an MQTT broker which is seen by HA as an integration
I try to learn the vocabulary at the same time …:)
Name is tydom2mqtt btw
An add-on is a very specific thing, "external docker" ... is not an add-on
Well how do you call an add-on when you use HA docker
I mean when you « install » manually something that HAOS calls an add-on
That's just software you're running using Docker
You could call it a container, if that floats your boat
NGINX addon is used to be able to access the DuckDNS domain while ussing HTTPS, is that right? I cannot figure out why it's not working.
The certificates are created by DuckDNS addon and they are there. NGINX runs and log shows no errors. Not sure what I'm missing. I can access the domain but only http, https fails to connect. :S Any ideas?
Do I need to open any port on my router? NGINX doc do not say anything about it.
#add-ons-archived have their own channel
And yes, you need to forward a port to NGINX... they probably didn't cover that as it's considered obvious
oops, I thought I was in #add-ons-archived , my bad. Thanks, it wasn't obvious for me 😦
Hello all! My ssl cert expired, can't access HA with duckdns. How can i connect to renew it?
Try the IP address. You may have to accept an invalid cert warning
i need to buy Raspberry PI and i tried to install Home assistant in VM under Windows and working fine with me but i did not see the supervisor option i tried to install " Studio core server Addons , it showing now but Addons not there
Are you looking in the right place? The Supervisor button in the sidebar was replaced by settings pages for Add-ons, backups, etc
I'd bet they're following a 2+ year old YouTube "guide"
The #add-ons-archived channel can help you find them, or press c and start typing add
hey all! I installed HA in docker on a pi. It seems HA doesn't come back up after a reboot sometimes... am I missing something?
There are times where it does come back up itself, but sometimes not.
Did you somehow manage to configure Docker not to start on boot?
I don't believe so. I have two HA instances at two locations, both cloned fully from the same image, with minor changes to IP address and whatnot as needed.
After doing apt update/upgrade and a reboot, one of the pis came back by itself, the other I had to run docker run homeassistant.
Then you should switch to using compose files
Ip address never worked for me after duckdns and ssl was configured. ☹️
Ohh, it worked in incognito window... (Duckdns address and accepting invalid cert)
Bizarre... I just tried rebooting again, and everything came back...
I guess I just gotta check after reboots to make sure it's back up and running.
If it doesn't check the logs to see why not, rather than simply restarting it
It's not that it fails to start because of an issue, it just seems like it doesn't even try to start HA after a reboot sometimes.
I'd hope that information on that should be there somewhere...
Well, I can't reproduce it right now, so I guess next time it happens, I'll try to dig in more.
Last question (hope so) after ssl renewal, some time need to pass for everything to settle? (Log says valid till mar 9) but still doesn't work
hello, quick question. I'm running HA on docker on a synology NAS. I plan to move it to a Raspberry PI 3 using HA Operating System. Is there a way to use my docker backup on the HA OS ?
You don't need a docker backup. Use the HA backup (which is really just the ./config directory).
But other than that, it should work OK.
does your rpi have a ssd? If you're using the sd-card, you're asking for headaches. Plus it'll be slower than your nas.
Depends on where the cert is being used, you may have to restart HA for instance, and you do have to restart NGINX
Not sure if this is a strictly installation question, but it's about backup/restoring an existing HA instance. I need to do a complete OS rebuild of the NUC HA (supervisor docker) it is running on. I have backups of HA but not entirely sure what's the best way to backup/restore zigbee2mqtt and avoid having to repair all my zigbee devices. I'm using Sonoff zigbee 3 USB dongle. I've seen Lewis's video but that was a year ago so really wondering what's the latest and best way to do it now that we are in 2023 🙂
If you're using the Z2M add-on then the backup for Z2M is included in the full backup
All add-ons are
While you're rebuilding, use HAOS instead of Supervised
Yes, that's exactly what I'm planning on doing. Supervised is giving too much grief
I saw the backup option and I've downloaded the zip. When I reinstall, when do I get the option to restore this backup?
If it's small you can do it before onboarding, but just don't
during the initial provisioning, you can restore rather than create user
Set up HA and then use the Samba add-on to add the backup, and restore that way
Z2M?
Restoring as part of the onboarding only works for small backups, and it silently fails otherwise
IIRC the ui for the restore is a bit... vague. It works but it's not always obvious when it's done.
my restores must be "small"
I have 65 z2m devices, is that 'small'?
It's the size of the backup that matters
Hey Tinkerer FYI I used your guide and managed to have my shell command working , thanks a lot :)
What zip file are you referring to? The HA backup is a tar
That's what she said
I'm talking about the Z2M backup. I wasn't sure whether a full HA backup would also contain z2m network and device pairings
Yes, it does
The full HA backup includes all the add-ons
Otherwise it'd be pretty damn useless
No harm in having the z2m backup but you shouldn't need it
Two obvious things missing from your guide:
- the public key must be added to authorized_keys (you mention to add the private key)
- the path /config/ssh doesn’t work for me. I have indeed a folder /config/ssh but I had to use the path ssh/ in my configuration.yaml otherwise key was not used
Oh wow. That's awesome. So a full HA backup is all I need. No need to be messing around with z2m backups and sonoff coordinator backup either then. Thanks!
Seems obvious but hey
Cheers
From the integrations page if I click ... next to an integration and select disable debug logging, it just downloads a log file and then never turns off debugging for the integration. Does anyone here have any idea what could be wrong?
How did you enable debug logging?
I don't specifically recall enabling it. I'm thinking it probably got enabled via default_config:
no, it will have sdcard. Is it a problem ? I consider switching from NAS to Raspberry for my usb dongle. Synology is messing around with usb driver on latest firmware...
Your setup will eventually die without warning if you use an SD card when the SD card inevitably fails
You can extend that timeframe by using an endurance or industrial SD card
If you're going to use a Pi though, use an SSD (on a powered USB hub)
you mean external ssd with external power ?
Yes
i didn't know those sd card were not reliable 😦
It won't fix the rpi being slow, but it will be less likely to just die on you
SD cards not being reliable is pretty well known
Pretty simple, use a 20$ usb 2.0 hub with dedicated power supply
Attach any hdd or ssd to it installed with raspbian
Boot from the disk
Enjoy
ok noted, thanks for the input
SSDs have controllers that manage the flash to do things like wear-leveling to extend their life. SD cards don't
and why an external powersource for the ssd ?
To avoid underpowering it and losing all your data 😉
okok
smh Sd cards don't stand up well to the write-heavy usage pattern of a database.
the pi has struggle to power an external ssd hence dedicated supply
You can use the hub to plug everything for your pi
Dongles, keyboard , etc
ok thank you all. so basically, if i copy my existing ./config directory onto HA OS, I should be fine.
Should I configure debug logging outside of default_config: if i'd like to disable debug logging for integrations from the integration page? Selecting an integration and clicking ... disable debug logging does not seem to work.
no, but the fact that you seem to have debug logging enabled for that component without doing it (or knowing that you did it) says that it's enabled in some other way, such as through the logger: integration
I specifically checked configuration.yaml and I have no entry for logger:
Also it's not just one integration that has the debugger enabled, it's all of them. Is there anywhere else I should check?
what exactly are you seeing that you want to stop?
I'd like to stop getting warnings and errors from the ReoLink cameras I have. (Logs found under Settings > System> Logs)
I'd also really like to know why the integration page shows the debugging icon for each integration I have enabled.
Last but not least I'd really like to know why I keep getting a system log file that is 30GBs in size.
I know it's an option to enable the logger in yaml and I could kind of control this from there...
that's not debug logging, and you can limit the logging for any component via the logger integration
what you were controlling was to make things more verbose, not less
what version of HA are you running?
I'm gonna guess that it's old, and that you have logger: in your configuration.yaml
Home Assistant 2023.1.1
Supervisor 2022.12.1
Operating System 9.4
Frontend 20230104.0 - latest
then the only explanation for a giant log is that you have logger: in configuration.yaml with debug or something
but you can look at your logs and see...
Hi ! so i just reinstalled home assistant yesterday. I checked that I could access the webpaage, I could, so I went to bed. Today, after coming back from work, I wanted to continue the configuration. But ican't access it anymore, and I'm accessing it vie homeassistant.local:8123. I didn't move the rpi, and I've tried rebooting it
Any clue what this could be ?
if it's just reolink, then you can change the log level yourself
Try using the IP instead of homeassistant.local
Mdns 🤮
doesn't work either
Try checking the console for errors
Also, what install method did you use?
but how ? with ssh ?
rpi imager
Plug in a keyboard and monitor
in the front, where the sd card is, there's a red led
and the ethernet led is blinking
it's a pi 3 without fans. could it be overheating?
Probably not, unless you have it smothered or sitting on other hot electronics
You pressed a key on the keyboard to wake it up right?
I think the monitor needs to be plugged into the Pi at boot to show anything
At least it does for me
Such as?
That's a good sign, you can troubleshoot errors 😂
You can share logs via a code sharing site
Please use a code share site to share code or logs, for example:
- http://pastie.org/ (select YAML for the language)
- https://dpaste.org/ (select YAML for the language)
- https://paste.debian.net/ (you guessed it, select YAML as the language)
Please don't use Pastebin, since it can randomly add spaces to the main view. Please also don't share text as images since it makes it harder for people to help you. Remember that others may have colour blindness, impaired vision, etc.
error -110 whilst initialising sd card
Sounds like a dying/dead SD card
but it has never been used 😦
Did you said you'd moved to SSD? Or was that someone else 👀
I never mentionned anything about it
Then I am confusing you with someone else, sorry!
No problem ! But could this really be a dying sd card? My father bought this 64gb card maybe 6 month ago, and never used it because I immediately "stole" it for the rpi
If you've not used it before today, it would be worth flashing it again and starting over
But that's what I did yesterday
I couldn't log to home assistant anymore, after letting the rpi unused for couple of weeks(and unplugged)
I took it back, couldn't connect, I reflashed, and I could connect, I went to bed, and now I can't connect anymore
So this already happened before where it just randomly stopped working?
well, starting 2 days ago, when I picked the pi back up
but before that, it was working "fine"
I think you may have gotten a faulty SD card then. It's definitely not unheard of. It's also more likely the SD card being bad than the pi
both have never been used before this
ill try getting another sd card then :/
16gb is enough right?
32GB is the minimum
If you plan to use an SD card as your main storage, you'll want a high endurance card
But if it's only 6 months old, contact the retailer to get it exchanged due to being faulty
How can I prove it's faulty? Because I wouldn't want them to put it in a camera and take a picture and say "look, its fine"
do you have a specific model in mind?
could it also be the sd card not being appropriate for this? I know, for instance, some sd card cannot be used with gopro
for whatever reasons
It won't be worth them arguing with you frankly, just tell them it's faulty and stored media can't be consistently recovered.
SanDisk or Samsung are both good brands, but there's other reputable brands also
its a netac 64gb, micro xc
Never heard of netac
me neither to be fair 😂 I wonder where he got these from
yes but any model would be fine?
Fell off the back of a truck?
You want high endurance, they'll say high endurance in the name or description
Well that would be the problem 😛
I think I have some Netac actually, I throw files on there for my 3D printer
yes, and they often come with a 128mb sd card haha
Please just make sure you make scheduled backups and copy them somewhere else. Even a high endurance SD card will die eventually and you'll lose everything if you don't have backups
thats honestly something I should do more often on my printers
I run both with klipper, so I have a lot to lose
It's a Linux thing iirc. Won't display unless it boots with a display
It definitely depends. I've had mixed results
Usually I can plug in a monitor no problem, but I've encountered it not showing anything on some machines
Ok so I setup Let's Encrypt and got the cert but how do I configure the OVA to use that cert and port 443?
More info is needed. What install method did you use, how did you get the cert, what's in front of HA?
Using the OVA for ESXi,
Let's Encrypt add-on to get the cert. Nothing is in front of HA. Local LAN
I'll do cloudflared for external access later
You can set the port to 443 and point it at the cert there. For the cert/key paths, use the paths that say "If using the Let's Encrypt add-on*
Thanks!
You'll also want to get the certificate expiry integration and make an automation to renew the cert and restart HA
Oh the Let's Encrypt add-on doesn't automatically check certbot every 30 days?
Nope, it never renews on its own
I just have an automation that tries to renew every day if it has < 30 days left
someone pointed out this: https://www.home-assistant.io/integrations/cert_expiry/
Hmmm does this use the let's encrypt config I setup or is it going to use it's own config? This has to be a DNS check as this server isn't exposed publicly right now
I think it just checks the cert
It looks at the cert yeah. Mine is not publicly exposed
has nothing to do with how it got there or what to do with an expired cert
It gave an error saying connection refused. Maybe since port 443 isn't open
I need to find a way to access the YAML for the OVA
yeah, it looks like you give it a host and it grabs the cert from there, rather than pointing it to the files
How do you normally edit the config?
You can use a vast variety of options lol
I'm just setting this up now and I was surprised that changing the port was not in the GUI
presumably you can access your host from home, so use that URL
Including vscode add on, SMB, file editor, etc
I can access the server but I don't see where in the UI to edit the raw config. I guess I can access the OVA directly but it gives me some stripped down terminal and not bash/zsh
Does this apply to the prepackaged ESXi OVA?
HAOS is the same once it's installed and running
Thanks. Trying to get HTTP configured with the certs now. Then I'll figure out the automatic cert renewal
I can paste my automation for it in a little while
Does it make sense to expose home assistant publicly for the app or use a VPN to connect home if you have one?
If you have a VPN and all of the people who use the mobile app also use that VPN, then I think it's best to just use the VPN
My home VPN is always connected on my phone
I don't want to be limited to vpn only, so I use a reverse proxy
Hi, I've been trying (and struggling) to install home assistant core on a clean copy of rasp pi (3b), and when I get to the final step of "install homeassistant" running into an error building cryptography. I've gone through their recommended debug steps and pretty stuck here. Any chance someone has a minute to prevent me from bruising my head against the wall?
What's the reason you're trying to install core?
As opposed to OS?
HAOS or Container are the two recommended install methods
Well I understand it (and I'm pretty new to this space, so forgive me) installing via HAOS won't allow me to keep other processes I want to run executing/accessable
I already have a codebase (python) I'm hoping to interface with HA. Admittedly I haven't looked much into containeer
Okay, so if you're looking to run other stuff then Container is the best option. You basically create a Docker Compose file, run the command, and you're up and running (after the slow pi fetches the Docker images)
Okay, I'll give that a try then. Embarrasingly, despite me being a software dev I've never run docker, but it's about time for me to learn!
Your head will thank you
Thanks for the tip!
I think I'm gonna expose it externally via cloudflared without any auth on Cloudflare's end
Thanks!
This is for the official Let's Encrypt plugin from here? https://github.com/home-assistant/addons/tree/master/letsencrypt
Yes
@odd marlin I suppose I might as well ask before I progress too much further: It is possible to interface with a HA dashboard (that is, push/manage data) programattically, using some rest/json/etc. API, yeah? As it stands, I'm only looking to use it as a self-hosted dashboard, but will probably later be looking to replace my current integration with Blynk and directly interface with my arduino devices.
What is your goal with the data?
Thanks. Looks like it will work. I guess we'll see in 90 days. Now to start adding devices to Home Assistant and getting some automations configured. I'm trying to slowly migrate everything from HomeKit to Hass as the brain, and then reconnect it to HomeKit for voice control
Currently my goal is just to get a data visualization dashboard (view sensor output history/statistics) up and running. For now, I'm content to run code/make decisions/communicate with devices myself. It seems HA is much more powerful than I have in scope now, but I'm good with incremental progress at this point
@odd marlin
getting this error
Failed to open a session for the virtual machine Home_Assistant.
The virtual machine 'Home_Assistant' has terminated unexpectedly during startup with exit code 1 (0x1).
Result Code: NS_ERROR_FAILURE (0x80004005)
Component: MachineWrap
Interface: IMachine {85632c68-b5bb-4316-a900-5eb28d3413df}
I don't think there's a way to modify the frontend with an API, but the underlying devices/entities/etc are exposed in the REST API
https://developers.home-assistant.io/docs/api/rest
You can also export metrics to things like InfluxDB, Prometheus, etc
I know nothing about Macs so somebody else will have to help you
Oh awesome, thank you. I've got a long way to go in my home management journey!
Best of luck. It's a fun journey
@placid rock are you running haos? in virtualbox or something?
Did you just install virtualbox? Did you reboot after install?
Disable virus checkers, if any?
how ?
@odd marlin Yo thanks for the help! Got it running in a container and I'm going through dashboard onboarding right now.
@placid rock no idea, I don't have any
@placid rock don't PM me man. If you have questions, say them here. I probably don't have much time to help.
I got a friend request too (since I have PMs disabled) that I rejected after I tried to help in #general-archived earlier
Hey all,
How do I set static IP for my VM machine (got one for HA and one for Plex)?
Best way is on your router. Should be able to assign static DHCP
Is there an alternative way to do that? I have no access to my router administrator panel...
You can assign static IPs per device. However that can cause things to become unstable, since if the router assigns x.x.x.5 to your phone and your HA vm is meant to have x.x.x.5, nothing will work
If you know the dhcp range, put it outside the range, but inside the subnet
eg: if your LAN is 10.1.1.* (/24) and the DHCP leases are 10.1.1.10 to 10.1.1.200; then make your static/stolen IP 10.1.1.201
Chances are they won't know the range if they can't access the router
Yeah I have no idea what the range is.
Plus a lot of consumer routers just do .5-.255
Or just .2-.255
I live in Israel and they charging money for accessing your control panel in the router there is a fee for DHCP static IP.. which is completely stupid...
Wait, so how do you change your wifi name or password?
A fee for that?
A fee for a static public IP is not abnormal. A fee to go to 192.168.1.1 and change your local LAN settings is abnormal
Can you get a new router
Or put your one into bridge mode
Or something
Or a new ISP
Yeah I think I will just get an open source router rather than this north Korea production 😂
A cheap PC from aliexpress running opnsense works well
I recently deployed one of these and it works great!
https://www.aliexpress.com/item/1005004603903448.html?spm=a2g0o.order_list.order_list_main.146.2cfd18029xkhaa
I run pfsense on my proxmox box. Works great
Hey, how to share integrated bluetooth from my computer (i have BT but i don't need a Bluetooth dongle to work) to a virtualmachine ?
As mentioned... you need to tell us which VM software you're using
OH sorry, Virtual Box + haos_ova-9.4
Is your Bluetooth from a wifi card? If so usually you have to pass the entire card to it
Set wifi card in "Network" tab ?
No, you'll have to pass through the card if that's where your Bluetooth comes from. Not set it up as a network
There should be a devices or similar name to pass through
I dont find it 😦
If I upgrade Home assistant OS, do any packages I install through the hass shell get wiped? I am thinking of installing CUPS to provide Airprint services on my HASS odroid.
you want to install that cups server with a addon, else you render your haos installation unsupported
got it. Ill ask in addons if there is one that works. Thanks!
Hello there!
Maybe someone knows what could go wrong
I got my server connected to my LAN and vice versa via wireguard, any port that I need to be forwarded - works, except for home assistant
I see it via nmap on 127.0.0.1, but it's not accessible from my LAN
Home assistant were deployed on docker, with host mode, also tried with bare port forwarding (just putting ports into compose.yml) - no results.
Firewall rules updated accordingly, for 8123/tcp on wireguard interface
I can provide you with screenshots of firewall configs, compose.yml, configuration.yaml and docker ps output too, but it seems to look fine (to me)
Nothing crashes, forwarding goes to 0.0.0.0:8123 and everything seems to work, but it's not as expected, huh
How/why is wireguard between your LAN and HA server?
127.0.0.1 is every machine's localhost IP - it does not leave or enter a host. It's the internal loopback address and is always pingable on that host, if TCP/IP is working correctly.
Forwarding to 0.0.0.0:8123 is not valid on a firewall/router. A router has to forward to one host a.b.c.d:8123
What's the LAN address for the host running docker?
I just don't have a Raspberry Pi 🙂
So I got a server on hetzner, my router connects to cloud VPN, providing me with access, and this server has access to my router LAN devices too
router -> hetzner; hetzner -> router
IP assigned to host in router's lan: 192.168.10.2 (pingable, accessible from router's LAN, services are usable)
192.168.10.2 is the host in the cloud via VPN?
HA not hosted on your LAN is not going to be very useful since many integrations do discovery via broadcast or multicast
Possibly complicates things, but... not very different from accessing HA remotely. HA would need full access to internet, and other containers/add-ons, which docker should handle
Oh, photos are forbidden there
Yes
In my LAN it's IP for host
"I see it via nmap on 127.0.0.1" - nmap from where?
That's pretty much a given then as it's just a loopback to itself
ss --tcp on host too
If it's not on the LAN, it's not going to discover devices and many integrations don't offer an option to provide the IP of the devices, so it's quite different from accessing it remotely
rule on host for firewall on wg interface has port 8123/tcp added and config reloaded
Yes, I can see the issues that might arise. The VPN is routing IP to/from the host and LAN but not likely UDP, etc, not without helpers.
But he's not even getting to step 1: http://192.168.1.100:8123 which should work, if configured right.
Okay setup for initial proof of concept but chock full of hazards and pitfalls for ongoing development or actual full blown usage of HA...
What IP does HA have configured?
Am I able to configure HA ip somehow? I just couldn't find such config
To discover local LAN devices it would need to be a bridged VPN, very likely in this scenario
So you deployed HA via docker?
Yes
And that's all so far?
Yep. Mapped volumes for config folder, tried to use ports: - "8123:8123" and then switched to just network_mode host, as in guide
Also tried to define external url in configuration.yaml as http://192.168.10.2:8123
It's been a while now but what's step 2 in installing HA? browse to ???:8123?
Seems to be, "Onboarding"
Besides docker config for compose.yml itself and firewall mapping that I performed further
Only if I didn't get it wrong and the guide has something else
Can you run this on the host?
http://<host>:8123
Yes. Ran. Gets HTTP 302, forwards to onboarding.html
If we take localhost as a <host>
