#add-ons-archived

1 messages · Page 60 of 1

sly badge
#

I can't see any published ports in portainer either not even for the web interface

viscid orchid
#

I don't use it, but it's not clear that you're expected to send data back to your instance directly, anyway. It looks like it logs into a cloud account?

#

right, it's disabled by default

sly badge
viscid orchid
#

what leads you to that conclusion?

sly badge
#

if there's no way to open port?

#

I'm just not sure how to open it

viscid orchid
#

I don't know anything about Traccar, but it looks like it's logging into an account, and not expecting you to forward a port to it

sly badge
#

as far sa i understand the tracker is not logging into account it just sends data to some port

manic fox
viscid orchid
#

this is logging into an account

#

Integrating into Home Assistant
The traccar integration of Home Assistant makes it possible to transfer all assets tracked by Traccar to appear in Home Assistant as a tracked device.

Add the following snippet to your Home Assistant configuration.yaml file.

device_tracker:

  • platform: traccar
    host: localhost
    port: 18682
    username: TRACCAR_EMAIL_ADDRESS
    password: TRACCAR_PASSWORD
manic fox
manic fox
sly badge
manic fox
sly badge
manic fox
#

just look at the HA config - it says default 8082 - so that port would be used for Traccar integration with HA

viscid orchid
#

except the addon tells you to use 18682, as above

sly badge
#

I don't need integration now. I need to send data to traccar from tracker that has it's own sim card to traccar

viscid orchid
#

ok, gonna bow out. I don't know how this thing works and it looks like it's going to take some research

sly badge
#

I just don't understand why there are no published ports in portainer next to traccar container

manic fox
#

if you are running as add-on - check the add-on logs - what does it say? and if using a add-on you should be able to open the UI from HA console itself. don't have to type it port

sly badge
#

It gives error

#

'HTTP 415 Unsupported Media Type - NotSupportedException (...)'

manic fox
#

you HA is supervised or Core?

sly badge
#

supervised

manic fox
#

so when you run the add-on the logs show the above error?

sly badge
#

Nope, that error only shows in the UI when I click on traccar in the left hand menu where all the other addons live

#

very frustrating that I can't send screen shots here 😅

manic fox
#

there is a log tab where you have the add-on - check that log if it has any errors

sly badge
#

Error loading shared library libsystemd.so: No such file or directory Error loading shared library libsystemd.so: No such file or directory Native library (linux-aarch64/libsystemd.so) not found in resource path (tracker-server.jar) - UnsatisfiedLinkError (... < HealthCheckService:40 < Main:111 < *:132 < *:106)

#

not sure that has any significance here

#

after restart of addon they went away

manic fox
#

of course it has - it did no install properly. Try reinstalling, it may not help though

#

you are on RPI or urbuntu?

#

I am just guessing you have a 32bit os typical raspian and this add-on might require 64bit os

sly badge
#

rpi4

manic fox
#

but raspian is installed as 32bit or 64bit? check with command uname -m

sly badge
#

Yeah it's 64

manic fox
#

just to confirm your uname -a gives results as aarch64 ?

sly badge
#

Linux a0d7b954-ssh 5.10.17-v8 #1 SMP PREEMPT Mon Jun 21 12:55:38 UTC 2021 aarch64 Linux

manic fox
#

okay - then try reintalling the add-on. as core libs are not found

#

before doing that

#

try the sudo apt update; sudo apt upgrade first

#

after that uninstall and re-install the Add-on from HA

sly badge
#

minor issue sudo: apt: command not found

manic fox
#

have you ssh into rpi as pi or equivalent access user?

sly badge
#

I'm using the ssh addon

manic fox
#

ah thats the problem - sudo access will not work there. - it is just a limited access add-on. Does not give complete OS acess

#

anyway just try reinstalling the traccar add-on and see if that logs and the error in UI goes away

#

I highly doubt it since this looks to the lib problem and some updates are needed - I am guessing you haven;t done the update and upgrade in long time

#

use putty like software to ssh into your pi and run those commands. You will anyway need it

sly badge
#

It won't connect

#

ssh just works with the ssh addon

manic fox
#

then you must be on HA OS - not supervised

sly badge
#

Sorry, I got a bit confused there it's the ha os

manic fox
#

just do a reinstall of the add-on than and try to see if it works. If not then I dunno - you might need to raise / check this in HA community

sly badge
#

Reinstalling doesn't really fix this 😦

#

looks like other ppl are having the same HTTP 415 error as well after core update

sly kindle
#

@sly badge do you have gps devices compatible with traccar for vehicles?

#

what's your intended use for it? Because there are easier ways to track people

#

especially if you are only using phones, anyway

river obsidian
#

Hoping this goes here instead of integrations?
I set up InfluxDB 2.0 with HA OS, and am fairly satisfied with the dashboarding on there. What are the reasons people use Grafana with HA vs using the built-in tools in InfluxDB? I literally just did this, so want to make sure I don't run into roadblocks with InfluxDB alone that then make we wish I had set up Grafana from the beginning

jade wing
river obsidian
#

Ok, will take this over there - thanks

sly badge
grim delta
#

Hello, I'm trying to update zwavejs2mqtt and get an error. Can you help me?

#

Can't install ghcr.io/hassio-addons/zwavejs2mqtt/aarch64:0.22.0: 500 Server Error for
http+docker://localhost/v1.41/images/ghcr.io/hassio-addons/zwavejs2mqtt/aarch64:0.22.0/json:
Internal Server Error ("readlink /mnt/data/docker/overlay2: invalid argument")

wild merlin
#

I have functions in node-red that I want to display on the dashboard when I access the dashboard via nabucasa… is this possible

formal furnace
#

Yes you need to access the site over https/SSL inorder for the barcode to work

wild merlin
formal furnace
#

??

#

Grocy? Idk

#

Haven't used nabucasa

wild merlin
#

Ahh ok, thanks

verbal valley
#

Hi everyone,

I am trying to develop custom add-on and having issues with ingress.
Basically trying to run calibre-web (https://github.com/janeczku/calibre-web) as HA add-on so I can access my e-book library anywhere.

Add-on works perfectly but when I set

"ingress": true,
"ingress_port": 8083,

in the config.json and try to access through HA. I get "Failed to load resource: the server responded with a status of 404 (Not Found)" but at the same time http://localhost:8083 works perfecty. So I assume server runs but just not sharing with HA.
I really do not know how ingress works.

Would anyone help me on this matter?

Thanks.

digital grotto
bright sphinx
#

hi all – desperately trying to setup unifi controller but I'm getting this error when I set-inform via ssh on my device: Status: Server Reject (http://192.168.1.193:8080/inform)

#

I've set the hostname/ip in the controller config and rebooted the addon

#

are there any other logs available any where?

#

I can ping 192.168.1.193 from the device

nova anvil
#

Hello everyone! Trying to use MotionEye node into NodeRed. Can someone tell me what paramerer name should I write to get the camera triggered motion detection into NodeRed?

#

In help it writes "Parameter name in motioneye transmission command."

#

But I cannot figure it out what is the transmission command in motioneye

bright sphinx
oblique lagoon
#

Is there any way to disable filewatcher on VSCode addon?

raven tinsel
#

Sure, you can configure what VSCode watches in the vscode settings

oblique lagoon
#

@raven tinsel I recall it didn't helped last time I tried it, but let me check it again along with some bigger scope of config_path

#

@raven tinsel So I set Watcher Exclude: /** and config_path: /.
After opening WebUI and closing it addon CPU usage is at 13%.

With config_path: /root looks better as CPU is at 0%.
I guess something is wrong with VSCode Exclude option

raven tinsel
#

So, why do you link the file watcher with CPU usage?

#

it works based on notifications, so its not a "scanner"

oblique lagoon
#

In my case leaving open VSCode Addon was resulting in "hearing" device from other rooms as fan was working like crazy on quite new Intel NUC (HASS is inside VM). Hence my thinking went at CPU side.

raven tinsel
#

Oh sure, I was not questioning your CPU usage in general, more the solution you are looking at feels weird, hence I'm wondering why you came to this as a solution for your problem

oblique lagoon
#

Been starting & stopping this add-on each time for a long time, hence started looking for clues.
CPU is readable within addon dashboard, difference was quite big to take a note here. I haven't touched CLI yet.
If I won't hear it anymore with config_path: /root and /** in exclusions I guess that's all from my side.

Another idea would be indexing, as this could explain high CPU usage.

raven tinsel
#

during startup that would be expected, but it should go down after that

#

if it keep "indexing", that is possible, I've seen that a couple of times

#

mostly that is caused by a circular symlink loop in the config folder

#

I've seen a lot of people having /config/config pointing to /config meaning there is a loop as it goes /config/config/config/config/config/config....

oblique lagoon
#

yeah, make sense

#

I will look for any symlinks in there

raven tinsel
#

I have no idea what on earth creates those loops on installation of people, but I've seen at least 10 cases of exact that

#

(over the past year or so)

oblique lagoon
#

I've run find /root -type l and looks good on my end. Anyway, thanks for your help and time you put into this community 🙂

worldly drum
#

Hello, I have a new device I would like to add support for using zigbee2mqtt with the addon for hassio, I see the documentation but am not super clear on some stuff, does anyone have a few minutes to help walk me through some of it?

willow lintel
# 
"/usr/src/homeassistant/homeassistant/components/hassio/ingress.py", line 120, in _handle_request
    source_header = _init_header(request, token)
  File "/usr/src/homeassistant/homeassistant/components/hassio/ingress.py", line 188, in _init_header
    connected_ip = ip_address(request.transport.get_extra_info("peername")[0])
TypeError: 'NoneType' object is not subscriptable
uncut hare
#

hey guys 🙂

I cannot start my esphome addon anymore. It says "Unknown error, see supervisor".
In the logs I got the following:

21-07-15 11:06:06 ERROR (MainThread) [supervisor.utils.json] Can't write /data/addons/data/a0d7b954_esphome/options.json: [Errno 2] No such file or directory: '/data/addons/data/a0d7b954_esphome/tmpbuv3xb4z'
21-07-15 11:06:06 ERROR (MainThread) [supervisor.addons.addon] Add-on a0d7b954_esphome can't write options

Someone knows why this happens? Had some problems with my mariadb password after update but I changed it (not anymore containing '@') and now everything besides esphome is working again.

#

Uhh i'm blind. Just found it on github, saying reinstall works. Let's try xP

#

works...^^

willow lintel
candid drum
#

When I try to install dnsmasq, I always get the error "Port '53' is already in use by something else on the host." Any solutions to this? Or alternative ports I can use?

viscid orchid
#

like this?

#

if you're running Supervised and not Home Assistant OS, you're probably already running a DNS server on the host. I don't see the port being in use with Home Assistant OS

silk marsh
#

Anyone able to help me with this one?
Can't install homeassistant/aarch64-addon-letsencrypt:4.11.0: 500 Server Error for http+docker://localhost/v1.41/images/homeassistant/aarch64-addon-letsencrypt:4.11.0/json: Internal Server Error ("layer does not exist")
I have already tried multiple reboots and ha supervisor repair

fallow finch
#

Whoops, I was in the wrong channel before:

Hi all 👋 I am at a loss with an issue. Suddenly my Ingress addons are all returning 401. Oddly enough, if I use the mobile app, it still works, just not when using the browser.

viscid orchid
#

are you connecting through a reverse proxy?

fallow finch
#

Yes

#

But I always have and it wasn't an issue before. The last thing that changed to my knowledge was an update to core-2021.7.2

#

I am seeing a lot of 21-07-15 21:05:38 WARNING (MainThread) [supervisor.api.ingress] No valid ingress session None

#

I have already cleared the cache for the domain btw,

#

And the problem persists if I try from an incognito window

fallow finch
#

Interesting, when I try another browser, it does work. So it seems to be a Firefox specific issue

unkempt dove
#

Hi. Can someone help me with NGINX / SSH?
I have a HA OS installation, with NGINX addon running for HTTP access from outside my internal network.
What I can't figure out is: if I want to SSH to my HA from outside - is port forwarding on my router (22 --> myhass:22) enough, or is there a way (and a use of it) to additionally configure my nginx for ssh traffic?

willow lintel
silk marsh
#

update on my problem with letsencrypt, although I don't think it's actually letsencrypt that is the problem

docker run --pull always homeassistant/aarch64-addon-letsencrypt:4.11.0

4.11.0: Pulling from homeassistant/aarch64-addon-letsencrypt
Digest: sha256:5638e17ff75fde349fe48fde72a4881de1b23802eccd75f97d26ef1eb437d617
Status: Image is up to date for homeassistant/aarch64-addon-letsencrypt:4.11.0
docker: Error response from daemon: layer does not exist.

That same command is fine on other systems and the command work with version 4.10.0 on the problem system, but I can't get docker to a stable state. It thinks the layer is missing if I try to run (or list, or delete) but thinks everything is already up to date if I try to download.

fathom bison
#

Let's Encrypt
errors runing the install any ideas with this message?

#

[17:01:50] INFO: Selected http verification
Requested name https://hapalmas.duckdns.org:8123 appears to be a URL, not a FQDN. Try again without the leading "https://".
[cont-finish.d] executing container finish scripts...
[cont-finish.d] done.
[s6-finish] waiting for services.
[s6-finish] sending all processes the TERM signal.
[s6-finish] sending all processes the KILL signal and exiting.

viscid orchid
#

did you try the suggestion?

fathom bison
#

😫 sorry jejej yes! now is working

#

well now I got times out...

viscid orchid
#

if you're using HTTP challenge, then you need to do this:

#
  1. HTTP challenge
    Requires Port 80 to be available from the internet and your domain assigned to the externally assigned IP address
    Doesn’t allow wildcard certificates (*.yourdomain.com).
#

are you using the DuckDNS addon? If so, you don't need the Let's Encrypt addon at all

fathom bison
#

ohh I thoug I need 2

#

both I meant

viscid orchid
#

no. DuckDNS takes care of the cert

fathom bison
#

OK let me uninstall them..

#

nop... when I write to config file I do have this error:

#

Invalid config for [http]: not a file for dictionary value @ data['http']['ssl_key']. Got '/ssl/privkey.pem\n-'. (See /config/configuration.yaml, line 14).

viscid orchid
#

looks like you have a newline there

#

review your formatting

fathom bison
#

man whas a underscore... I did not notice,!
Thanks A lot!!

dusty hatch
#

Hello. New here. Am looking to get support for zwavejs2mqtt add-on.

ERROR (MainThread) [supervisor.api.ingress] Ingress error: Cannot connect to host 172.30.33.0:8099 ssl:default [Connect call failed ('172.30.33.0', 8099)]

Zwavejs2mqtt 0.22.1
System: Home Assistant OS 6.1 (armv7 / raspberrypi4)
Home Assistant Core: 2021.7.2
Home Assistant Supervisor: 2021.06.8

Please and thank you!

copper gazelle
#

Fresh install of hassos on raspberry pi. Installed samba share first thing. Connected just fine using my macbook. However, when I attempt to move my snapshot, it yells at me and states: "Modifying 'homeassistant' requires an administrator name and password. To move 'snapshot.tar' click Authenticate"

viscid orchid
#

Is the user that you're logging in with an admin?

copper gazelle
#

Yup, I just figured it out. Check the logs, dummy!

#

my mac was attempting to use my old account, because it had it saved.

viscid orchid
#

it's never quite clear what requires "admin" and what requires "owner" - it's all very much a work in progress

copper gazelle
#

It's all good. Love it enough to deal with a few minor issues here and there

umbral raptor
#

Anyone have any idea why the Node-Red integration is not in the list for intergrations? After the update it wouldn't load and I removed it and can't re-add it
Solved (I think)

brittle mortar
#

This first response was this:```This can be a problem with the overlay storage.

Try running ha supervisor repair from the terminal. Please note, this command can take a bit to run.```

weary stratus
#

Hi there, is there still an issue with the latest Unifi Controller AddOn? Since the last three updates it remains "offline" or not reachable. After restarting the supervisor I cannot connect to my Unifi controller anymore. I could bypass by deinstalling the AddOn, re-installing the AddOn importing the last backup. This would last until the next restart of Home Assistant. Acually the Unifi Controller log says

org.tuckey.web.filters.urlrewrite.UrlRewriteFilter INFO: destroy called
Exception in thread "Thread-15" java.lang.IllegalStateException: BeanFactory not initialized or already closed - call 'refresh' before accessing beans via the ApplicationContext
    at org.springframework.context.support.AbstractRefreshableApplicationContext.getBeanFactory(AbstractRefreshableApplicationContext.java:177)
        at org.springframework.context.support.AbstractApplicationContext.getBean(AbstractApplicationContext.java:1123)
            at com.ubnt.service.ooOO.Ö00000(Unknown Source)

Any idea?

unreal dirge
#

Since i went 64bit the Mopidy add-on from Kent Loving does not work anymore. I tried one made by Poeschl instead, and while it does install and play TTS i can't make it play wav or mp3.... Maybe someone knows about a nice alternative, or maybe even another way of playing both TTS and wav/mp3 through the output jack?

vocal juniper
#

Good morning.

peak lance
#

I want to run AdGuard on my HA Blue. Can i somehow run Unbound with that?

silk marsh
weary stratus
#

Hi there! So after updating controllerx (appdaemon), I no longer can use the 'arrow' buttons (< and >) on my IKEA remote, to change bulb color/temperature 😦 Anyone know what can cause this? The logs show:

ValueError: This light does not support xy_color or color_temp
Which isn't true. I can change the color_temp manually through HA ui. And it worked fine a couple of weeks ago.

cyan zenith
#

Hi all, just having some issues integrating my unifi controller into Home Assistant, getting conflicting reports on access via port 8443 and getting no help from the service logs, anyone able to lend a hand? Cheers 🙂

fathom bison
#

Hello Fello HA!

fathom bison
#

I having issue to connect from outside using.. the android app!

sly kindle
#

@fathom bison what are you using now?

fathom bison
#

Nginx Proxy Manager

#

@sly kindle with duckdns.org was working fine but when I was in my house could connect wifi to the app. so I read using Nginx Proxy Manager could solve the problem but now nothing work no outside or inside (the app) if I going http: Local with a browser works.

fathom bison
#

anyone has idea how connect the app internally on you lan network with Nginx Proxy Manager

#

?

viscid orchid
#

I can't really parse your description of your current configuration or what's working/not working

#

NGINX Proxy Manager works fine for me everywhere

#

this whole sentence was confusing:

#

with duckdns.org was working fine but when I was in my house could connect wifi to the app. so I read using Nginx Proxy Manager could solve the problem

#

what problem?

fathom bison
#

Hey @viscid orchid sorry let me clarify, duckdns was working fine... later I realize when I arrived home the app did not worked internally with my wifi connected. I research and find out that my router do not do loopback. So I need to look another option found Nginx Proxy Manager (proxy server) turn off duckdns off and keep with Nginx Proxy Manager. but still I am in the same place I connect internally with wifi home and the app do not work...

viscid orchid
#

you can just use a separate internal_url that connects directly

#

or maybe not with SSL configured

#

that's where NPM would help

fathom bison
#

but then my connection won't be secure..

viscid orchid
#

on your network?

#

correct, but who cares?

fathom bison
#

nop outside of my network...

viscid orchid
#

NPM/NGINX supports a secure connection from outside, but your local insecure connection is still available when you're on your local network

fathom bison
#

in theory... but when i am going through the app does not work

viscid orchid
#

you can use http://xxx:8123 URL for internal_url and (probably) set your Wifi SSID to tell the app to use it

fathom bison
#

you are right internally i do not need ssl.... and I do that in the ap wifi:xxxxx and source http://

viscid orchid
#

question

fathom bison
#

is very specific on the app in android... but do not work...

viscid orchid
#

if you're using NPM for SSL, then you need to disable it for HA

fathom bison
#

humm.,, where in the config fiel?

#

file?

viscid orchid
#

for what?

fathom bison
#

disable it for HA?

viscid orchid
#

your /config directory

#

you must have edited it to enable SSL through DuckDNS

fathom bison
#

looking for that le me see...

viscid orchid
#

from the DuckDNS addon docs:

# 
Additionally, you'll need to configure the Home Assistant Core to pick up the SSL certificates. This is done by setting the following configuration for the HTTP integration configuration in your configuration.yaml:

http:
  ssl_certificate: /ssl/fullchain.pem
  ssl_key: /ssl/privkey.pem
#

then you need to pay attention to this:

fathom bison
#

ok i am confused, did you run duckdns and also Nginx Proxy Manager ?

#

or is one or the other?

viscid orchid
#

I don't use DuckDNS

#

I use a different DNS provider, but it looks like NPM supports DuckDNS cert challenge

fathom bison
#

yes I did the install and works fine but my question do you also have to have the dnsduck plugin install and runnig?

viscid orchid
#

no

#

it looks like NPM supports DuckDNS cert challenge

#

that's what matters

fathom bison
#

ok could be not working them because I not doing DNS challenge ... I thing that is the problem...

viscid orchid
#

you have to set up the cert in NPM

fathom bison
#

oh yeah the cert is set up...

#

that is whay I can connect from outside...

viscid orchid
#

if all that is working, then comment out the ssl_* lines and follow the instructions in the breaking changes: <#add-ons-archived message>

#

and if it's working through NPM already, maybe you already did all that.

#

so how are you trying to connect to your HA instance from your local network?

fathom bison
#

yes my only is with the app on my phone....

#

when I got home the app does not work I have to turn of the wifi...

viscid orchid
fathom bison
#

I did but does matter give an error .... also the ssid of my wifi...

viscid orchid
#

can you use that in a browser on a PC connected to your local network?

fathom bison
#

that is my error... if I go to my laptop and connect wifi from the browser works fine...

viscid orchid
fathom bison
#

OK thanks

#

I will shoot something there.... hey @viscid orchid thanks a lot for your time and help!!!

ruby spire
#

sorry this is a nooby question but how do i access my HA remotely? I tried with duckdns but am only able to access locally. How do I fix this?

viscid orchid
#

did you forward port 443 in your router to your HA instance?

paper bluff
#

I am using conbee with aqara sensors. I moved to new hardware and now my automations don’t work. How can I confirm that there is connection between the conbee stick and aqara door sensors?

#

I am using node red for automation

dense spadeBOT
#

Please DO NOT cross post. Read the channel description, post it and wait for folks to respond.

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.

Command '.crosspost' was invoked by: tom l
peak lance
#

Is there a way to reach an addon outside the HA menu?
I'm trying out AdGuard and i want to reach it without having to go through HA.
It's basically for using AdGuard Sync and it needs access to it.

vocal juniper
#

Is there a posibility to connect via the WireGuard module to my Root Server? My Router isn't accessable from the internet so it can't use the forwarding - however I can use my Rootserver as WG Server. But I didn't found documentation how to make a WG Client on HASS

viscid orchid
vocal juniper
#

I have no clue how to integrate this in HA

#

I tried to add it as Repo - but it fails and also into HACS

#

what's the trick/documentation?

ruby spire
#

I tried doing 8123->8123 and 443->443, 443->8123

#

Still unable to access from outside network

viscid orchid
#

How are you accessing it locally?

viscid orchid
#

I don’t use it, so you’re on your own as far as how it works

vocal juniper
#

thx a lot - looks great now

ruby spire
#

These are the ports that I have forwarded

viscid orchid
#

http://homeassistant.local:8123?

#

and how are you trying to access remotely?

ruby spire
#

8123->8123 and 8123->443

#

I’m trying to use my duckdns domain to access

viscid orchid
#

if http://homeassistant.local:8123 is working, then you haven't enabled SSL

#

did you follow the instructions in the DuckDNS addon?

ruby spire
#

my bad

#

i restored to an older snapshot because i had broken something

#

but that only worked locally

viscid orchid
#

works now locally?

#

or some time in the distant past?

ruby spire
#

not anymore

#

i reset to an old snapshot before i did any of the duckdns

viscid orchid
#

so redo it

#

I go back to: Did you follow the instructions in the DuckDNS addon?

#

you at least need to get back to the same functionality you had before

ruby spire
#

ok let me try to get back

ruby spire
#

i think there is something wrong with my portforwarding

#

i checked canyouseeme.org and put in the two ports and it couldnt find anything

viscid orchid
ruby spire
#

no i couldnt

viscid orchid
#

actually, that's wrong too

#

should be: https://xxxx.duckdns.org, which uses port 443

ruby spire
#

i followed the steps and followed a youtube tutorial didnt work

viscid orchid
#

you should follow the addon instructions, not a YT video

ruby spire
#

i did both

viscid orchid
#

it's wrong

ruby spire
#

well then im kind of stuck

viscid orchid
#
  1. Follow the DuckDNS addon instructions
#
  1. Forward port 443->internal_ip_address_of_ha:443
ruby spire
#

so i dont even need 8123?

viscid orchid
#

8123 is the non-SSL port

#

you can tell it to use any port you want, but the default is 8123 for non-SSL and 443 for SSL

ruby spire
#

i contacted my ISP and they said that they only allow port forwarding for ports 25 and 80

#

does anyone have a workaround for this?

viscid orchid
#

You can forward one of those 2 ports to 443 on the HA server and use :xx to specify the port when you connect

ruby spire
#

ok that makes sense

#

i just realized another thing

vocal juniper
ruby spire
#

my isp gives me a modem with a built-in router and i have my own personal router connected to that router

viscid orchid
#

You should put it in bridge mode

ruby spire
#

so would i have to port forward from my isp router and then again from my personal router

viscid orchid
#

That’s a mess

ruby spire
#

what should i put in bridge mode?

viscid orchid
#

The isp router

ruby spire
#

what if i dont have that option?

viscid orchid
#

Good luck

ruby spire
#

would my other approach work tho?

viscid orchid
#

You can try

#

Or you can just use Nabu Casa

ruby spire
#

alright thanks

vocal juniper
#

^^^boradlink IR "learning" IR codes?

# 
Learning...
260050000001269214111113141114111113141114351213153412381237143512371435131213361311143512131213143514111114141111381311143512371411143512371435140005250001284811000d05
Base64: b'JgBQAAABJpIUERETFBEUERETFBEUNRITFTQSOBI3FDUSNxQ1ExITNhMRFDUSExITFDUUEREUFBEROBMRFDUSNxQRFDUSNxQ1FAAFJQABKEgRAA0F'``` omfg
ruby spire
#

@viscid orchid if I already have it in bridge mode what would I have to do to make sure portforwarding works

viscid orchid
#

Then it’s just your router

ruby spire
#

When I go on can you see me it says connection refused?

#

Is that a problem with NAT loop back?

viscid orchid
#

no, it has nothing to do with that. you're asking the website if it can connect to it

#

mine succeeds

ruby spire
#

Oh

#

My god

#

I finally fixed it

#

So for anyone who may have the problem later you have to portforward from your isp router/modem and your own router

fluid rune
#

wait you're using two routers?

ruby spire
#

My ISP provides me with kind of a 2-in-1 modem/router so im forced to have 2 networks if i want to use my own router

#

So to answer your question, yes i am using 2 routers but my personal router is connected to the modem/router my ISP gave me

stoic swan
#

Is HACS the same as the Community Addons in the Supervisor?

#

and if not (which it doesnt look like) what is the reasoning behind the two different stores?

#

is HACS an option still when running HASS OS?

digital grotto
#

hacs does not have addons

#

and yes hacs is an option when running hass os 🙂

stoic swan
#

but what does hacs then have? i guess there is terminolgy thing there

digital grotto
#

integrations and ui stuff

stoic swan
#

ah.. so they are not installed separately (as the addons are) but right in to core?

digital grotto
#

yes

stoic swan
#

ok. i guess i will have to try it 🙂

#

thanks

pastel plinth
#

Hi. I'm having trouble setting up emulated_hue. I have Home Assistant running on rPI (connected via cable). I've used Home Assistant OS for setup. When I restart everything I get this lines in log

Traceback (most recent call last):
  File "/usr/src/homeassistant/homeassistant/setup.py", line 249, in _async_setup_component
    result = await task
  File "/usr/src/homeassistant/homeassistant/components/emulated_hue/__init__.py", line 133, in async_setup
    listen = create_upnp_datagram_endpoint(
  File "/usr/src/homeassistant/homeassistant/components/emulated_hue/upnp.py", line 73, in create_upnp_datagram_endpoint
    ssdp_socket.setsockopt(
OSError: [Errno 99] Address not available
#

I'm trying to bind it to port 80 (which is needed for Alexa)

vocal juniper
#

By using Wackeup on Lan and SSH i managed to put my Windows PC in hibenate and wake it up - just with a HA swich ... "going to turn down the PC" evening loop optimized

cunning blaze
#

Is there a way to install an older version of an Add on? I want to install an older version of Z-Wave JS

viscid orchid
#

restore it from a snapshot

willow lintel
#

hi, using Nginx since yesterday, please help me find an ipv6 address that gets logged with an invalid authentication

#

ipv6 is disabled on my router, and I am not aware any integration is using it, so how can this be?

mystic mango
#

Greetings all

#

I have HA running on VirtualBox and I have been using mosquitto add on, but I am not sure how I can access the mosquitto container from CLI. I am trying to clear the retain messages so I can start from scratch

#

I guess I would remove the add on and start over, but really don't want to do that if I don't have to.

#

I used to run mosquitto on a different server but since moving I figured I would try the add on

#

now just not sure how to access the mosquitto db

vocal juniper
#

@nova tendon I didn't get it how to run docker cmds .. even with ssh enabled

nova tendon
#

the only way that i can get that is on directly 2nd console of hassio

vocal juniper
# 
-bash: docker: command not found
nova tendon
#

because the ssh instance is already a docker container that dont expose the docker.sock

vocal juniper
#

check that's the explaination

nova tendon
#

so you must get access to the host ...

vocal juniper
#

what I didn't understand

#

we have 2 SSH add-ons .. why and what's the difference?

#

ok I can now access docker from the terminal

quaint estuary
#

Has anyone here used "NGINX Home Assistant SSL Proxy" along with DuckDNS for external accessm and suddenly gotten a "400 bad request" when trying to access your home assistant? It began suddenly and has been working for months

fathom bison
#

I am still fighting trying to configure... local access with android HA app. WIFI when I arrived home from outsite working fine... not luck yet....

solar locust
#

NGINX requires configuration changes to HTTP: in configuration.yaml. Check the release notes.

viscid orchid
main cargo
#

For the Unifi controller add on, I am getting a "ERR_SSL_VERSION_OR_CIPHER_MISMATCH" error. Any ideas on how to solve? This used to work fine for about a year

solar locust
#

Have you restarted the add-on? "Due limitation, renewed SSL certificates are not picked up automatically. You'd have to restart the add-on in order for UniFi to pick up the change." Other than that, my controller is working fine today.

main cargo
#

I have, yeah. Just recently upgraded to 0.23.2

#

I'll try rebooting the entire server

solar locust
#

I just upgraded also. no problems. Are your other SSL-related add-ons running?

main cargo
#

DuckDNS is, yeah

#

Reboot did not resolve 😦

solar locust
#

Are you able to turn off SSL in the controller configuration and successfully open the web UI?

main cargo
#

Nope, says

"Bad Request
This combination of host and port requires TLS."

quick trench
viscid orchid
#

Not true anymore 🙂

#

The Web Terminal
You can access the web terminal by clicking the "Open Web UI" button on this add-on's Info tab. If you set the "Show in sidebar" setting (found on the same Info tab) to "on", a shortcut is added to the sidebar allowing you to access the web terminal quickly.

#

Home Assistant Add-on: Terminal & SSH

stark grove
#

Also just noticed that SSH & Web Terminal only has rating 1 now?! Whats up with that? Should I reinstall the Terminal & SSH instead?

viscid orchid
#

if you want

#

it's mainly about what you enable, and I think the drop was explained somewhere

#

you should use the one that does what you need

stark grove
#

Aha. Ok, well I don't suppose I use the web terminal anyways. As the latter is the official addon perhaps that would be the best, if equal SSH-wise.

viscid orchid
#

Well, you get the web terminal with both, as I posted above. But agreed that using anything with additional capabilities that you don't need, don't use, and maybe don't understand will just get you in trouble

stark grove
#

Agreed. Can always upgrade if there should be a need for it. Thanks 🙂

vocal juniper
quick trench
#

yeah, no, you just installed SSH & Web Terminal if you have a web terminal

viscid orchid
quick trench
#

nevermind

#

apparently that's new?

vocal juniper
#

but "Terminal & SSH" has ssh an a web terminal 🙂

#

I have HA for 3 days

viscid orchid
#

it's been there for a number of months

vocal juniper
#

no idea what "new" is 🙂

quick trench
#

frenck should just remove the community one then

#

new as in, newer than the last time I looked at the integration

viscid orchid
#

it has way more stuff

#

Docker access, for one

vocal juniper
#

ok that's nice

quick trench
#

well where were you 5 minutes ago when I answered derharry

#

tsk tsk rob

#

dropping the ball

vocal juniper
#

I asked the strange docker question - I guess times

viscid orchid
#

3mins late

vocal juniper
#

and it made no sense - but somebody explained that ssh is acually also a docker

viscid orchid
#

"container", but yes, every addon is a container

quick trench
#

all addons are docker containers

#

apparently the community ssh has super access

viscid orchid
#

the SSH & Web Terminal addon has a large list of features in its docs

quick trench
#

I don't know why I'm using supervised anymore then

potent lichen
main cargo
solar locust
#

I'm feeling lucky then. Had no issues with the update.

potent lichen
#

hmm it seems its not offering any TLS - tested with testssl

#

@solar locust do you see this in your logs? The JKS keystore uses a proprietary format. It is recommended to migrate to PKCS12 which is an industry standard format using "keytool -importkeystore -srckeystore /usr/lib/unifi/data/keystore -destkeystore /usr/lib/unifi/data/keystore -deststoretype pkcs12".

#

and whats your last message in the log ?

solar locust
#

Yes. That error is expected and can be ignored.

#

'''

#

Importing keystore /tmp/tmp.48Lwq48QXO to /usr/lib/unifi/data/keystore...
Warning:
The JKS keystore uses a proprietary format. It is recommended to migrate to PKCS12 which is an industry standard format using "keytool -importkeystore -srckeystore /usr/lib/unifi/data/keystore -destkeystore /usr/lib/unifi/data/keystore -deststoretype pkcs12".
[cont-init.d] unifi.sh: exited 0.
[cont-init.d] done.
[services.d] starting services
[services.d] done.
[04:33:37] INFO: Starting the UniFi Controller...'''

potent lichen
#

thanks ... same here I guess it just doesnt start ...

quaint estuary
solar locust
#

@quaint estuary YW

twilit idol
#

Does the LetsEncrypt add on automatically renew the certificates?

potent lichen
#

@main cargo did you make any progress? the way I see it the unifi service starts and stops right after

main cargo
dense spadeBOT
weary stratus
#

I am not able to configure duck dns. I have followed 2 articles(including official documentation) , each had different method,but non worked.

#

I know this is incomplete info, but even i dont know what i need to provide so pls ask me. To say the least, i configured port forwarding first, the i configured duck dns and registered a domain and then i configured duckdns on hassio. I also edited the configuration.yaml file.

solar locust
#

We'll need to see your configuration files and details of your setup.

quick trench
weary stratus
# quick trench all you need is a port forward from 443 to 8123, the ddns account & token, and a...

Ok, So, 1) I have port forwarded external port 443 to internal port 8123. I have entered the token and ddns. The logfiles of duckdns seems to be good. I will be attatching them additionally.
and 2) I have added these lines in configuration.yaml.

http:
    base_url: https://#my Domain#.duckdns.org
    ssl_certificate: /ssl/fullchain.pem
    ssl_key: /ssl/privkey.pem

I read on the forum that base_url has been depreciated so instead i need to enter my internal ip and external domain in configuration>general.

  1. I have a question, do we need to write the port 443 in the external domain field in duck dns addon configuration? though In either case, it was not working but its better to ask.
dense spadeBOT
weary stratus
main cargo
#

Still no dice with the "ERR_SSL_VERSION_OR_CIPHER_MISMATCH" error with the Unifi Controller Add On 😦

I did these steps so far:

  • Tried changing to use no SSL
  • Uninstalled and re-installed the DuckDNS add on
  • Uninstalled and re-installed the Unifi Controller add on

Nothing seems to work, I continue to get this error.

potent lichen
#

@main cargo I am in the same boat - I went further, revert to backup (unfortunately was not old enough) looking at the container it seems neither mongodb nor unifi starts but no real indicators I believe we need to wait for @raven tinsel to issue a newer version and until then not be able to manage the environment

main cargo
#

To clarify, the add on starts for me and the logs look fine, I just get the SSL issue when accessing the page.

potent lichen
#

it "looks" like it does but it actually doesnt the underlying services arent starting

#

if you go into the container and do a service mongod or unifi status you will see they arent running

main cargo
#

Whoa, I just got it! Uninstalling the Unifi controller and then making sure SSL isn't checked worked!

#

this isn't a full solve, since I cannot use it with the DuckDNS provided cert though

potent lichen
#

? you uninstalled it and unticked ssl after the reinstall?

main cargo
#

yep

potent lichen
#

I beleive I tried that didnt seem to help either

quick trench
#

in your configuration for ddns, you also need to list your url under domains as yourdomain.duckdns.org

domains:
  - yourdomain.duckdns.org
#

lastly, you'll have to use that URL as your internal url as well.

#

lastly lastly, the internal url won't work if your router doesn't support nat loopback. So to test properly, you have to use an external connection to verify it's working.

potent lichen
#

@main cargo you were actually correct - seems like the services dont recognise in the container which is weird ... I should be able to see if a service is running

lime goblet
#

Do you use firefox?

#

You need to disable the enhanced tracking protection for the Home Assistant website.

#

No idea why it recently changed, but that was the issue for me.

#

You can do it with the little shield in the url bar.

hidden delta
#

anyone use Samba Share. I am trying to connect it to my Synology Disk station. But it states "Cannot access share. It seems that your configured share does not exist." but then it shows the share as one of the shares the software finds.

# 
    ---------       ----      -------
    HA-Backups      Disk
#

with my target in the config as host: 192.168.1.2 share: HA-Backps target_dir: HA-Backups

potent lichen
#

I use Samba Backup 4.5.0 which works for me

#

samba share just shares HA folders via samba

hidden delta
#

sorryt yes i am using samaba backup

potent lichen
#

can you post your entire config - except password/username ?

dense spadeBOT
potent lichen
#

ok

#

change to target_dir: DailyBackups

#

HA-Backups is already exposed via the share name

#

if that doesnt work put the HA-Backups into quotes "HA-Backups"

hidden delta
#

oh that lookes like it worked

potent lichen
#

yours was pointing to target_dir: /HA-Backups/HA-Backups/DailyBackups

#

so that obviously didnt exist

hidden delta
#

ok i understand now. thank you.

potent lichen
#

no worries happy to help

#

I keep 4 local on my HA and 2 on the nas

hidden delta
#

if you wanted to put in in the root folder do you just target_dir: '/'

potent lichen
#

correct

#

oh depending on your setup this can take a while - mine takes ~90mins for an entire backup

hidden delta
#

mine took about 5 min

potent lichen
#

haha nice

#

mine is 6GB

hidden delta
#

ah im at 400mb

potent lichen
#

7.63GB actually

hidden delta
#

well thank you for getting me all set. feel much safer with some off system backups

potent lichen
#

definitely I used google backup before which was ok

#

but i rather keep it local

peak lance
#

My NodeRed addon web ui is not working anymore.
Get "401: Unauthorized".

Run HA Blue. Supervisor log say WARNING (MainThread) [supervisor.api.ingress] No valid ingress session None

lime goblet
#

Do you use Firefox?

peak lance
#

Yes

lime goblet
#

You need to disable enhanced tracking protection for your home assistant webpage

#

you can do it in the url bar with the little shield

peak lance
#

New thing since? Like now?

lime goblet
#

no idea. I had the problem since last update

peak lance
#

I use a hardened Firefox. Let me check.

peak lance
lime goblet
#

its not a nodered issue btw, just incase you think that. its a problem for all addons

willow lintel
#

just started with Nginx and all is working (well, got a silly error during HA startup, but other than that) fine. Please allow me this fundamental question: why is it better to use unsecured local traffic in the first place? Why are we more secure with Nginx, than we are using local secured traffic without Nginx

#

given the fact it is way more complicated to configure, and we have many addresses to trust, I am tempted to say a lot more can go wrong without exactly knowing what to do...

#

And I dont experience any speed advantage on the http traffic at all, which was one of the other side-effects I was hoping for

sly kindle
#

@willow lintel I don't think anyone says it is better. If you want to secure local traffic you need to take extra steps

#

Split brain DNS, or something else

willow lintel
#

ok thanks. I installed it for the HA ios app to be able to connect locally over the network and not via my duckdns address

fathom bison
willow lintel
#

first did that with adguard home (no special settings required) but then moved to Nginx because it was core add-on versus community

sly kindle
#

That happens because the cert is only good for the domain. In a browser you can continue past a warning about the cert when you enter https://local_ip:8123 But the apps won't offer this prompt and just fail stop there.

willow lintel
#

needs some extra settings though, all explained in the http: documentation

willow lintel
fathom bison
#

so you recomend adguard?

willow lintel
#

well, since I dont graps all details yet, Ill leave the recommendations to Cogneato 😉 can only say it was easier to configure. Kind of like the idea even my local traffic is secured over https

fathom bison
#

wow, I just figured out I was using Nginx Proxy MANAGER... so I will stop that one and give a try to the NGX assistant SSL proxy...

viscid orchid
#

you can add most types of customization to NPM as well

sly kindle
#

@fathom bison NPM is just a way of running Nginx

#

Basically web browser ui management vs a text file

viscid orchid
#

it can be a little annoying figuring out what UI box goes in which section, but you can insert stuff pretty much everywhere

sly kindle
#

yeah (the little annoying part)

fathom bison
#

Thanks, both!

willow lintel
#

Hmmm going back from nginx to adguard somehow won’t allow the local network anymore…

nova tendon
#

hi

#

my node-red is getting info from HA but the triggers not work

#

what can i do to check?

placid lotus
#

What is the easiest way to wipe all topics in MQTT and have it start fresh?

viscid orchid
#

MQTT Explorer

brazen grove
sweet cargo
#

Hello all, question regarding Hass Blue (Odroid N2+). Is this running 64bit (or 32bit) of Hass OS? Reason I ask Unifi 32bit addon is no longer developed. I was about to migrate my Unifi to Blue, but first would like to be sure this thing is running a 64 bit version of Hass OS? System health does show CPU Architecture: aarch64, but I'm not sure if this is just saying the CPU is capable of 64bit or whether this is saying I'm running a 64bit OS. Thanks!

#

@raven tinsel hopefully you know?

raven tinsel
#

aarch64 = 64 bits

sweet cargo
#

Cool, thanks!! Nothing stopping me from migrating then.

#

Is there a way to tell the Unifi addon to use the management vlan to reach unifi devices and the rest of hass & addons to use my IoT vlan?

willow lintel
#

had it working before without changing anything, out of the box so to say, but now cant get it to work, so figure that could be it

pure cipher
#

i try to connect vscode to home assistant. a few weeks ago it works but now i get a fault code
Connecting to Home Assistant...
[Auth phase] Initializing WebSocket connection to Home Assistant
[Auth Phase] Connecting to Home Assistant... Tries left: 3
[Auth phase] Received a message of type auth_required
[Auth phase] Received a message of type auth_invalid
[Error - 11:27:24] Request textDocument/completion failed.
Message: Request textDocument/completion failed unexpectedly without providing any details.
Code: -32603

and in home assistant i get a notification
Login attempt or request with invalid authentication from xxx.xxx.xxx.xx See the log for details

does anybody now what i can do to let it work?

willow lintel
random coyote
#

I just updated my Home Assistant install, and nginx which has been working fine since November is throwing a 400: Bad Request error on trying to load my main HA domain. It however is happy to load my subdomains that I have. Eg. abc.duckdns.org doesn't work, but unifi.abc.duckdns.org does work. What am I missing here?

viscid orchid
#

Did you review the breaking changes for HA 2021.7?

random coyote
#

Oh.... that's probably a good idea... LOL

viscid orchid
#

Yeah

random coyote
#

Do you have an example of what I should be adding? I read the docs, but yeah...

#

the trusted proxy is my HA install? Or my whole domain?

viscid orchid
#

It’s your proxy server

random coyote
#

So I'm using the nginx add-on so that would just be the local IP of HA?

viscid orchid
#

No

#

First, the logs tell you the (current) address

random coyote
#

Ah!

viscid orchid
#

Second, you should use the network shown in the http: docs that contains it rather than just that address

random coyote
#

So I can just use the address in the docs?

#

So the IP shown in the logs is just my internal gateway IP 192.168.0.1

viscid orchid
#

The error that says it’s rejecting your incoming request?

random coyote
#

[21/Jul/2021:10:53:46 -0400] - 400 400 - GET https xxx.duckdns.org "/service_worker.js" [Client 192.168.0.1] [Length 16] [Gzip -] [Sent-to 192.168.0.18] "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.114 Safari/537.36" "https://xxx.duckdns.org/service_worker.js"

viscid orchid
#

Not that one

neon yarrow
#

Hi. I'm super-new to this stuff, and I have some really dumb questions. I really did try to Google some of this stuff, but didn't find answers that were helpful to my situation... At least not ones that a dunce like me could understand...

Today I received my CC2531. It was pre-flashed with Z-Stack Home 1.2. I installed Home Assistant on my Raspberry Pi 4. The interface is working well. I added the Zigbee2mqtt add-on and figured out that I also needed to add the Mosquitto broker add on. Now I'm getting this...

random coyote
dense spadeBOT
neon yarrow
#

That's two halves of the same log message. Sorry.

viscid orchid
#

Please stop with the code walls

viscid orchid
neon yarrow
#

Okay, sorry.

random coyote
neon yarrow
#

Ignore. I'll go to the MQTT Discord and see there.

hollow ether
#

Asking another MQTT question: How do I connect an external MQTT client to the Mosquitto MQTT installed using the Configure->Integrations->Add->Mosquitto in HA? I.e. how do I add a login to MQTT running inside HA (HA is using the HA .vdi running on virtualbox).

#

What I'm trying to do: I have an external raspberry pi running RTL_433 that is listening to Accurite refrigerator sensors. I can read the temperature already on the raspberry pi. I want to send that information to my Home Assistant VM.

viscid orchid
#

the MQTT addon uses HA users by default, so the easiest thing to do is to create a special "mqtt" user for this

#

all the normal MQTT ports are already exposed, as shown in the Configuration->Network section of the addon

#

you've only mentioned the HA integration that talks to a Mosquitto broker somewhere, but I'm assuming that you're also using the Mosquitto addon. If not, then this question is more for generic Mosquitto configuration

hollow ether
#

The only thing I see in the configurations window of the add on is "Broker: core-mosquitto" "Port: 1883" "Username" homeassistant" "Password:". There is no place for me to add in addition users or configurations.

viscid orchid
#

the addon?

hollow ether
#

Yes.

viscid orchid
#

that's the addon

hollow ether
#

Ok, I'm lost, where do you get that menu?

viscid orchid
#

in the addon

dense spadeBOT
#

An add-on is a Docker container with tweaks to allow it to be configured in the Supervisor UI. It is only available in the Supervised installs (#330990055533576204 and #330944238910963714) since if you've used any other install method you're able to install software already.

Command '~addon' was invoked by: RobC
#

Integrations integrate Home Assistant with devices or services, or provide functionality within Home Assistant. Add-ons provide additional software or services, which an integration could possibility integrate with. Add-ons are for Home Assistant OS and Supervised only, other install methods can install software other ways.

Command '~integration' was invoked by: RobC
hollow ether
#

Never Mind. Found it under Supervisior -> Configuration.

viscid orchid
#

the broker is an addon, the integration talks to it

hollow ether
#

It wasn't under configuration -> Mosquitto-> configure

viscid orchid
#

that's the integration

#

it's under Configuration->Integrations

#

but your question is about other clients talking to the broker, and the integration has nothing to do with that

hollow ether
#

Yes, that's what I'm trying to do. I just assumed that since I installed it under the integrations that it would be configured under that tab.

viscid orchid
#

you could use any broker anywhere, even cloudmqtt

hollow ether
#

Yup. I understand that. Just got confused because I had installed it under integrations (which I guess installs the broker).

viscid orchid
#

addons are only available for Supervised installations (OS or Supervised), so HA doesn't configure them

#

installed it under integrations (which I guess installs the broker).

#

no

#

it installs the integration that can talk to any broker anywhere, including the addon. if you want to use the addon, you have to install that yourself (unless things have changed)

#

it's been a while since I installed the integration, and there's at least one example of an integration installing an addon (zwave_js), so maybe it does that now

hollow ether
#

Oh, did I get that wrong? That Mosquitto broker is installed by default as part of the OS/VM? So... technically I don't need to install the integration? Or, wait, I would still need to install the integration in order to subscribe to the messages.

viscid orchid
#

to keep it simple:

hollow ether
#

I think the problem that threw me off is that under Integrations is shows "Mosquitto broker" -> Configure, which makes me think that I can configure the broker under this block.

viscid orchid
#

the Integration, as defined above, allows HA to subscribe to and publish messages to a broker, anywhere. doesn't matter which broker it is or where it is, as long as HA can log into it. It has nothing fundamentally to do with the addon

#

the addon is an instance of a broker that you can optionally install and connect to HA via the integration.

hollow ether
#

Gotcha. Thanks!

viscid orchid
#

HA used to have a built-in MQTT broker long ago, but it's long gone. That made it even more confusing for a while

hollow ether
#

Ahhh....

#

I see.

candid drum
#

Since I installed duckdns, none of my integrations work anymore. (For example my robot vacuum cleaner)
I think I have to replace the internal IP addresses in my config file. Since the duckdns instance can't reach my internal IP's?
Am I right? And what should I change them to.

TL;DR Changed my baseurl to something.duckdns.org, now my local ip products don't connect anymore. (E.g. 192.168.1.135)

#

Found the solution myself. DuckDNS uses SSL (https) and my internal IP's don't. So i added http:// to the internal IP's.

junior saffron
#

Hey everyone. Ive been digging for hours now but cant seem to solve a problem I have. In short, I have influxDB running in my home assistant. Its working great with HA and has been logging everything perfectly.

Of note HA is a VM on a small linux box I have. On that same box I have 'unifi-poller' running and its working EXCEPT when it trys to input data into influxDB.

The error seems to be with dns. Given influxDB is running in HA, what ip should I tell external sources to 'publish' data to in order to access the 'HA/influxDB' instance?

#

PS: What other info can I provide to clear any q's up?

restive echo
#

anyone have issues with HAOS in a VM and the ESPHome add-on not being able to resolve mdns correctly?

red quarry
#

Anyone installed their own Almond server? I am getting some oauth errors when adding the bone assistant skill…

#

I end up with an error -
Sorry that did not work
Missing OAuth Client Secret in Authentication part of the manifest

#

This was when I used the almond server add-on.

#

Similar error when I run it in docker for windows…

prime parcel
#

Anyone having issue with addons library

#

21-07-22 08:15:47 WARNING (MainThread) [supervisor.store.data] Can't read /data/addons/git/a0d7b954/traccar/config.json: required key not provided @ data['boot']. Got None

#

I' m getting the same error for all the addons

prime parcel
#

Anyone?

willow lintel
#

because my instance wont come back on my iphone after updating to 2021.7.4 I checked supervisor logs, which is all green, except for 2 warnings I cant understand

#

[supervisor.addons.options] Unknown option 'host' for MariaDB (core_mariadb) and [supervisor.addons.options] Option 'interface' does not exist in the schema for Samba share (core_samba)

#

is this to be ignored? cant find anything on these 'options' in the add-on documentation?

sly kindle
#

@willow lintel if you reload default for those addon configs you'll see that the options are gone.

#

Alternatively, you could remove the specific options

raven tinsel
#

interface has been removed

willow lintel
#

thanks! so do we need to replace the host: '%' from before with anything now?

raven tinsel
#

this is not related to host?

#

just the interface option is removed (and now auto detected)

willow lintel
#

the host option is from MariaDB

sly kindle
#

There is no host: option

#

Check the docs for the addon

#

There's databases: logins: and rights:

raven tinsel
#

Correct, has been removed as well

#

(long time ago)

willow lintel
#

yes, and in both, I had the host: '%' setting

raven tinsel
#

That is at least a year ago

willow lintel
#

o dear... didnt notice this at all

#

thanks, will clean up immediately!

#

still had grant: ALL PRIVILEGES ON too....

#

but that didnt show a warning in the logs 😉

raven tinsel
#

it was granted 😬

willow lintel
sly kindle
#

I'd think that's related to your ssl config or what you have been messing with recently?

#

Weren't you bouncing around with nginx and Adguard etc?

willow lintel
#

no, thats was on my other (production) instance, which is working perfectly fine, (btw without the Nginx and/or adguard which I have uninstalled)

#

if I check certificate upon the warning, I see an incorrect duckdns.org domain listed. it should probably be deleted, but I have no idea how? The domain is configured correctly in Duckdns add-on

sly kindle
#

Reset all the things

willow lintel
#

thats a bit generic 😉 which things where? got to realize it is only iOS devices that have the issue. I just updated all devices, restarted everything reloaded Duck dns add-on (indicating all is fine and no renewal is needed)..

#

even manually deleted the tokens in the .storage, and have them recreated.. upon login, I see the token being added, but it still leads to a notification of unauthorized login attempt

willow lintel
#

I can see the certificates in the Mac keychain app, and managed to delete them There were long expired certificates there too I see now. And still the (now deleted) certificate shows up... this is true horror

sly kindle
#

@willow lintel yeah I was referring to site settings in browsers, certificate files

#

I don't know. Is it safari which still shows the deleted certificate?

#

I would ignore the HA companion app for now

willow lintel
#

not sure. I deleted the certificates form the keychain, and yet they still popup on the certificate warning. On Dekstop we can ignore that, but when I do that on Ios (both Chrome and Safari) i am not authenticated

sly kindle
#

DNS?

#

you cleaned up after your attempts with nginx and so on?

willow lintel
#

this is not the instance with Nginx. I had never changed this at all. Maybe I should go back to a purely local instance over http, and then rebuild ?

sly kindle
#

yeah but nginx and adguard are for your whole network, typically

#

you don't set them up per instance

#

you set them up anywhere on your network and they effect your whole network

willow lintel
#

yes, I understand that, thanks. Yet I can confirm that it worked just fine after I had deleted those. things started to go wrong yesterday, after updating to 2021.7.4

#

this is 1 instance of 3 behaving erraticly

raven tinsel
#

but honestly, I have no clue

willow lintel
#

first things first 😉

sly kindle
#

I would try reverting back to local http, uninstalling duckdns, clearing out the /ssl directory, reboot, add it all back fresh

willow lintel
#

ok will do, report back, if and when. thanks for your patience.. appreciated!

sly kindle
#

did you clear safari cache on these devices?

#

including site data

#

I don't know where else they would be pulling the out of date duckdns info

willow lintel
#

ok success! here's what I did. deleted the keys from the /ssl folder. Once again deleted the certificate form keychain Mac os. Uninstalled Duckdns. (myDesktop instance was still live while doing all of this). I then re-installed Duckdns letting it re-create the keys. After that, I could log back in without an issue.

tiny forge
#

This started happening yesterday:

Failed to start add-on
404 Client Error for http+docker://localhost/v1.41/containers/create?name=addon_a0d7b954_zwavejs2mqtt: Not Found ("No such image: ghcr.io/hassio-addons/zwavejs2mqtt/aarch64:0.23.2").

Is this happening for anyone else? Seems like it can't resolve ghcr.io but I can ping and curl the URL just fine

All my other add-ons are working fine.

viscid orchid
#

that's a generally a DNS problem

#

could also be lack of space to download the new image

tiny forge
#

@viscid orchid that's what I was thinking but I can ping and curl just fine from terminal. Disk wise DF shows only 37% of disk is used.

viscid orchid
#

try switching your DNS to 1.1.1.1 or 8.8.8.8

sly kindle
#

@tiny forge the terminal from the addon will show that the addon can ping and curl

tiny forge
#

@viscid orchid Changed DNS to 8.8.8.8 and no love

sly kindle
#

And rebooted?

tiny forge
#

@sly kindle will try that

#

@sly kindle after reboot and checking nmcli that DNS is set to 8.8.8.8 still getting same error.

#

If I need to reinstall is there anywhere that I can grab the network key for zwave since I can't start the UI?

sly kindle
#

snapshots?

tiny forge
#

tried that, no luck

sly kindle
#

what do you mean tried that

tiny forge
#

I tried restoring the snapshot

sly kindle
#

so you've already reinstalled?

tiny forge
#

no... I haven't just tried restoring

sly kindle
#

which addon are you using

#

when I go to the zwavejs addon config the key is right there

tiny forge
#

Z-Wave JS to MQTT
Current version: 0.23.2 (Changelog)

#

My other instance back on 0.22 is working fine.

#

@sly kindle when I launch the terminal (protection mode disabled) and check the folders the "addons" is empty. Is there somewhere else I should be looking.

sly kindle
#

that folder is for addons you would develop yourself or download manually

#

it's empty for most users

tiny forge
#

makes sense.

sly kindle
#

to test from within a running container you would need to use docker commands, which you can do with protection mode off

tiny forge
#

ok, doing a docker ps the zwave one isn't running

sly kindle
#

docker exec -it homeassistant /bin/bash would get a shell within the homeassistant container for example

#

you would check supervisor container. get in it, see if it can resolve

#

I'm installing zwavejs2mqtt. never tried it

tiny forge
#

So I did a docker pull for that image and now it's starting just fine. As if the image wasn't really downloaded even though it showed up in "docker images".

#

Thanks for the help!

willow lintel
patent copper
#

My ESPhome addon is suddenly failing to start with this in the log:

21-07-23 20:49:14 ERROR (MainThread) [supervisor.utils.json] Can't write /data/addons/data/a0d7b954_esphome/options.json: [Errno 2] No such file or directory: '/data/addons/data/a0d7b954_esphome/tmp9g_bp73t'
21-07-23 20:49:14 ERROR (MainThread) [supervisor.addons.addon] Add-on a0d7b954_esphome can't write options
21-07-23 20:49:24 ERROR (MainThread) [supervisor.utils.json] Can't write /data/addons/data/a0d7b954_esphome/options.json: [Errno 2] No such file or directory: '/data/addons/data/a0d7b954_esphome/tmp6izkmeep'
21-07-23 20:49:24 ERROR (MainThread) [supervisor.addons.addon] Add-on a0d7b954_esphome can't write options
#

Running on latest HAOS. Any ideas?

sly kindle
#

@patent copper whenever I see "can't write" I uninstall and reinstall the add-on.

#

don't know if that will help here though

patent copper
#

Did the trick! 😉 Thanks!

#

Should have thought of trying that myself really....

ionic harness
#

Anyone having an issue or found the solution for Nodered not showing debug logs or states of sensors?

onyx sandal
#

so i installed the lets encrypt add on months back, and everything was working fine. my cert expired, and now i can't get back into my home assistant interface. i thought it would auto renew for some reason

#

thats half true. i cant do it from my computer but i can get in from my mobile phone.

sly kindle
#

@onyx sandal restart the addon

#

And check the docs for automating renewal

onyx sandal
#

it was not started when i logged in, so i clicked 'start'

sly kindle
#

The let's encrypt addon docs that is

onyx sandal
#

so its running now

sly kindle
#

@onyx sandal Ah you're right. I thought it used to

onyx sandal
#

hmm the logs say challenge failed

#

but the dns record points to the device, http challenge should succeed.

sly kindle
#

Do you have port 80 forwarded

onyx sandal
#

yeah i mean this was all 100% 24 hours ago, today my cert expired and now its not working

#

let me look at the router configuration

#

i have external port 443 forwarded but not 80

sly kindle
#

Youll need 80

onyx sandal
#

do people open 80 and then close it again after cert is renewed?

sly kindle
#

Apparently people do 😃

#

You can do that, but it needs to be open for the challenge

onyx sandal
#

i must have i guess

#

at least i'll remember next time that i did this to myself

#

presuming this works

#

ya that worked

#

thanks cogneato

sly kindle
#

You can use that to show and/or notify you

mellow whale
#

im having issues installing DuckDNS. I was getting error codes but now its just the spinning circle over the word install and nothings happening. I was able to install the File Explorer add-on a couple days ago just fine tho

#

finally gave me an error code again
Can't install homeassistant/aarch64-addon-duckdns:1.13.0: 500 Server Error for http+docker://localhost/v1.41/images/create?tag=1.13.0&fromImage=homeassistant%2Faarch64-addon-duckdns: Internal Server Error ("Get "https://registry-1.docker.io/v2/": net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)")

sly kindle
#

Probably DNS. Try changing it to 9.9.9.9, 1.1.1.1, or 8.8.8.8

mellow whale
#

is there a way to change it thru the GUI? or do i have to do it thru the command line on the host?

sly kindle
#

@mellow whale Supervisor > System > Host card click on "change" across from your IP address

pure egret
#

Hey!

Would love some help with DuckDNS

For some reason that I don't understand, it doesn't work for me. Here are the steps I took:

  • Created address reservation for the HA VM
  • Port forwarded the HA internal IP and (service port: 433 & internal port 8123)
  • Downloaded DuckDNS and configured token & domain & accepted terms
  • Configured .yaml like set in the documentation and added Base_url

Would love for some assistance - the log from DuckDNS seems good, I don't know what the issue is.

Thanks!

sly kindle
#

@pure egret base_url was deprecated long ago so I assume you are using out of date videos or another guide instead of the docs?

#

You generally described the correct steps, however. What is the result?

pure egret
#

Hey @sly kindle, thank you for your reply!

I went over it and deleted the base_url, for some reason I can only reach my HA with my internal ip and not through my duckdns

Any ideas?

sly kindle
#

Which are you able to connect with? http:// or https:// ?

#

If it's http:// then there is a problem with your configuration.yaml

#

Using the local address

pure egret
#

I log in with https://, http doesn't work

mellow whale
#

@sly kindle changing the DNS didnt work. Do i have to change ipv6 stuff with it? I tried 1. 8. and 9. with ipv4 with ipv6 set to DHCP and disabled and nothing worked.

sly kindle
#

@mellow whale no. You would only need to choose static, leave everything else as is except for the DNS server. Then reboot

#

You could disable ipv6 but that's probably not related

mellow whale
#

ok

#

my dns was already 8.8.8.8, but i changed it to 1.1.1.1 and rebooted and still getting same error

pure egret
#

@sly kindle I think you may have found the issue

Unfortunatly I can't share screenshots here, but when I checked my IP on https://www.portchecktool.com/ this is what I got

"Problem! I could not see your service on My_IP on port (443).
Reason: Connection timed out."

Does it mean that I'm not port forwarding correctly?

I verified and the port forwarding is using the right internal ip adress (the one I log with to HA and I have adress reserved)

fiery prism
#

Either that or your ISP is blocking outgoing ports.

pure egret
#

Thanks,

Happen to know of a guide that I can make sure I port forward correctly?

I will call my ISP tomorrow and see what's happening on that front as well.

mellow whale
#

any other ways to get DuckDNS working? cause i really wanna use presence detection but i cant without the external ip

fiery prism
mellow whale
#

any ideas how to fix this? Can't install homeassistant/aarch64-addon-duckdns:1.13.0: 500 Server Error for http+docker://localhost/v1.41/images/create?tag=1.13.0&fromImage=homeassistant%2Faarch64-addon-duckdns: Internal Server Error ("Get "https://registry-1.docker.io/v2/": net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)")? havent really found anything useful on forums and i cant use HA for what I want until i can install addons

raven tinsel
#

That indicates a network problem

#

DNS is not resolving

mellow whale
#

i know that, but thats literally the only info i've gotten and still no idea how to actually fix it

raven tinsel
#

that all the info Home Assistant has as well

#

it connect and the connection fails 🤷‍♂️

mellow whale
#

so i just cant install add-ons until i find and fix a mysterious issue?

raven tinsel
#

It seems so

#

there is nothing HA can do about not being able to access the network/internet

mellow whale
#

i was able to install the file editor add on a few days ago and havent changed any settings since

#

this is the second Pi that ive tried to install it on, and ive tried ethernet and wifi on both and nothing. guess i just cant use HA

#

fun

raven tinsel
#

I'm sorry to hear you are having network issues

#

It has issues connecting to https://registry-1.docker.io/v2/, which is Dockerhub where all the running images are coming from

mellow whale
#

idk how that helps lol

raven tinsel
#

Well, IDK what you'd expect us to do, the device is not allowed to connect to that host from your network connection you have provided the device

mellow whale
#

i mean, was hoping someone had an idea on where to start cause idk what im doing. guess ill try re-installing HA again. idk what else to do

raven tinsel
#

This is not an HA issue

#

re-installing should not resolve it

mellow whale
#

ive tried changing the DNS settings multiple times and nothing changes

raven tinsel
#

can be anything, DNS, adblockers, firewalls, routing issues on your provider end, unstable connections

mellow whale
#

That sucks that I literally just can’t fix it since I have no info to go off of. Cant really use HA for what I want then

pure egret
sly kindle
#

@mellow whale did you reboot the whole thing after setting DNS or just restart Home Assistant?

onyx olive
#

Can I mount my NAS SMB share through the SSH & Web Terminal somehow?

mellow whale
#

I clicked the reboot host button. My dns was already set to 8.8.8.8 tho so I don’t think it’s that

sly kindle
#

@mellow whale have you tried restarting your router too?

#

Reboot any switches etc

mellow whale
#

ya. I mostly needed to install add-ons to get an external url/remote control working for geo-fencing, but i ended up just using HA Cloud instead since i dont need any add-ons for that

paper bluff
#

I updated to the latest version it broke my reverse proxy. I am using letsencrypt add on. THis is my config

  ssl_certificate: /ssl/fullchain.pem
  ssl_key: /ssl/privkey.pem
  use_x_forwarded_for: true
  trusted_proxies:
    - 127.0.0.1
    - 192.168.10.0/24```
#

what am I missing?

crimson gull
#

This issue doesn't just occur with this repository, I think nearly every repository I want to add doesn't work. For example Spotcast (https://github.com/fondberg/spotcast) also does not work

fiery prism
crimson gull
fiery prism
#

Spotcast is an integration not an add-on.

fiery prism
gilded canopy
#

oh my bad, thought it belonged here, let me move it 🙂

copper dagger
#

Does the add-on mosquitto broker ever just soft-ban a device? I have tried restarting the broker and the whole raspberry pi and the device that has issues. Currently the device connects but no subscriptions or publishes go through

#

It is in fact an esp8266 with a ir code sender sketch

#

To add a bit more the device was working for a while and then randomly stopped doing anything besides "connecting"

#

A reset back to a backup worked Ignore my previous post

junior pebble
#

Do we need to add anything into yaml for MQTT to work?

junior pebble
hardy mulch
#

any idea if there's a way around my problem?
Home Assistant with 2 network interfaces, one on my home VLAN, the other on the IOT VLAN where all smart devices are connected.
If HA primary interface is set to the server VLAN, some integrations break (TP-Link Kasa) because it seems they only look for devices through the primary interface.
If HA primary interface is set to IOT VLAN, the Samba add-on since 9.5.1 can only be reached through HA primary interface, which for security reasons I cannot access from my home VLAN directly. Versions before 9.5.1 allowed you to specify an interface to expose SAMBA on, but this option has been removed. I don't think there's a way to downgrade an add-on

balmy elk
#

Hi together, im using hassio / ho & nginx proxy addon to redirect requests to different services. After the last update there was a bracking change in core-2021.7.4 during the use of reverse proxys. https://www.home-assistant.io/integrations/http#use_x_forwarded_for . Sure i read all i could find, but i never get HA Frontend reachable again from my external domain - no matter what i tried to customize in my configuration.yml. Always get a "400 bad request back." Could someone help me out / gimme a hint what i'm doing wrong with my proxy config?

#

this is the ugly message: A request from a reverse proxy was received from 172.30.33.8, but your HTTP integration is not set-up for reverse proxies

#

And thats ma http settings:

#

use_x_forwarded_for: True
trusted_proxies:
- 172.30.33.0/24

balmy elk
#

just tried it like so:

#

use_x_forwarded_for: true
trusted_proxies:
- 172.30.33.0/24
- 172.30.33.8
- 127.0.0.1
- ::1

#

i must miss something here, thats for sure 🙂

balmy elk
#

use_x_forwarded_for: true
trusted_proxies:
- 127.0.0.1
- 172.30.33.0/24

#

also no 🙂

raven tinsel
# 
http:
  use_x_forwarded_for: true
  trusted_proxies:
    - 172.30.33.8
atomic drift
#

greetings everyone

#

im having trouble getting snapshots from motioneye throught the snapshot url

#

im getting 401 unauthorized

#

even with _username and _signature in the url

#

because it's missing the cookie as i've observed in the network tabs

raven tinsel
#

use the integration? :S

atomic drift
#

but should i get the cookie somewhere or is there another way around it

#

i need to get it from another computer

#

for a report i build every week

raven tinsel
#

that will not work via ingress

#

in that case open it up in the add-on network settings to an outside port (and use that)

atomic drift
#

ok

#

the motion_webcontrol port ?

raven tinsel
#

oh right

#

not need to open a port

#

its already open by default

#

motion_webcontrol?

atomic drift
#

i was lookin at the documentation, yeah that one right

#

its not bound by default on my config, bounding it right now

raven tinsel
#

Yep, it is closed by default indeed, as a general good practice idea

atomic drift
#

nice

#

thanks 🙂

balmy elk
#

@raven tinsel the base variant with only the Proxy ip - did not work for me, too. restartet the whole machine

#

use_x_forwarded_for: true
trusted_proxies:
- 172.30.33.8

#

A request from a reverse proxy was received from 172.30.33.9, but your HTTP integration is not set-up for reverse proxies
14:34:50 – (FEHLER) HTTP - Die Nachricht ist zum ersten Mal um 14:34:17 aufgetreten und erscheint 9 mal

raven tinsel
#

Running Supervised? 🙂

balmy elk
#

it looks like its ignoring what i hack in - always the same message 🙂

#

hmmm

#

hei hfunny a new ip

atomic drift
#

why not allow the whole range ?

balmy elk
#

ok thi xould not work 🙂

raven tinsel
# atomic drift why not allow the whole range ?

A reverse proxy basically can spoof an IP. By specifying a range you are basically saying: Any IP in this range, may fake the client address.
That is something you should not (unless that whole CIDR is used for a cluster of reverse proxies)

balmy elk
#

oki that could not work, i wanna say 🙂

#

but its a bit surprising. look at the first post, there ist was the 172.30.33.8

atomic drift
#

i thought the range being local 172.30.33.0 meant that until someone ran a rogue container on his network it would be safe to allow the whole range

#

my bad if it was plain bad advice :X

balmy elk
#

first try was with 172.30.33.0/24 also not working. (c-class range wildcard, if i understand right)

raven tinsel
#

That should just work (eventhough I would not advise it)

#

if not, you are having different problems

balmy elk
#

well i have different problems, that why im wondering/writing 🙂

#

because whatever i have tried so far - nothings working like before with the proxy 🙂

raven tinsel
#

A request from a reverse proxy was received from 172.30.33.8, but your HTTP integration is not set-up for reverse proxies -> Means your config is incorrect

#

that simple 🙂

#

So you are probably setting it on the wrong place

balmy elk
#

is it? 🙂 happy for advice

#

worked since years without problems 🙂

raven tinsel
#

Well

balmy elk
#

missing what is wrong atm

raven tinsel
#

you had it configured wrongly for years in that case

#

or else it would not have been broken now

#

So from a security perspective, you had an issue

#

which we now prevent

#

so did it work for year?

#

No

balmy elk
#

mire than a year...

raven tinsel
#

yeah it did work, but was incorrect configured

#

that is now detected

#

we just saved your butt from a wrongly configured system with potential security issues due to misconfiguration

balmy elk
#

great

#

but, how to fix it? Settings in the proxy addon?

raven tinsel
#

nope

#

Configure your HTTP integration:

http:
  use_x_forwarded_for: true
  trusted_proxies:
    - <PROXY IP>
#

that is all that is needed

balmy elk
#

sry i dont get what you mean actually.

#

here is what i configed and what the log says

#

A request from a reverse proxy was received from 172.30.33.9, but your HTTP integration is not set-up for reverse proxies
14:50:30 – (FEHLER) HTTP - Die Nachricht ist zum ersten Mal um 14:45:24 aufgetreten und erscheint 6 mal

#

use_x_forwarded_for: true
trusted_proxies:
- 172.30.33.9

#

this should work, shouldnt it?

raven tinsel
#

Please wrap you code in triple back ticks

#

I cannot tel[l if you identation is correct now.

#

additionally, I don't see if you are setting it on the http configuration

#

as the context around that is missing

atomic drift
#

so if i understand correctly, changing from 172.30.33.0/24 to the actual ip of the nginx container would be a safer setting ?

#

(for my case where i allowed the whole range /24)

balmy elk
#

''' # Uncomment this if you are using SSL/TLS, running in Docker container, etc.
http:
base_url:*
server_port: 8123
cors_allowed_origins:
- https://*
- http://*
use_x_forwarded_for: true
trusted_proxies:
- 172.30.33.9
ip_ban_enabled: true
login_attempts_threshold: 5
api_password: !secret api_password '''

#

Sry try to get used to discord...

sly kindle
atomic drift
#

` use those

# 
code
atomic drift
balmy elk
# 
# Uncomment this if you are using SSL/TLS, running in Docker container, etc.
 http:
   base_url: ******
   server_port: 8123
   cors_allowed_origins:
     - https://*****
     - http://*****
   use_x_forwarded_for: true
   trusted_proxies:
     - 172.30.33.9
   ip_ban_enabled: true
   login_attempts_threshold: 5
   api_password: !secret api_password
sly kindle
#

@balmy elk 3 backticks, paste your content, close it with 3 backticks

shadow cairn
#

Hi. I get this in gitt pull add-on: [Error] git origin does not match https://github.com/...
what does it mean?

sly kindle
#

should be seeing that in logs

balmy elk
#

alright, commented it out and actually do a host reboot

# 
# Uncomment this if you are using SSL/TLS, running in Docker container, etc.
 http:
   #base_url: ******
   server_port: 8123
   cors_allowed_origins:
     - https://*******
     - http://********
   use_x_forwarded_for: true
   trusted_proxies:
     - 172.30.33.9
   ip_ban_enabled: true
   login_attempts_threshold: 5
   api_password: !secret api_password
#

host back online,

#

now lets see if its reachable 🙂

# 
A request from a reverse proxy was received from 172.30.33.9, but your HTTP integration is not set-up for reverse proxies
15:10:47 – (FEHLER) HTTP - Die Nachricht ist zum ersten Mal um 15:10:43 aufgetreten und erscheint 8 mal
sly kindle
#

@balmy elk I just noticed the space in front of http:

#

is that how it is in your config? if so, then none of that config is being used

balmy elk
#

well THAT could be a could explanaiton 🙂

sly kindle
#

this is why it is important to included the whole thing and using formatting here in discord, so those details can be seen

raven tinsel
#

base_url and api_password don't belong in that config

sly kindle
#

I explained about base_url, was going to get to api_password next 😛

#

this is like some ancient config, or copied and pasted from somewhere 👴

balmy elk
#

like things like this happen... ill try without the space 🙂

#

dont ask how long im sitting here searching for a cause, not seeing this...

sly kindle
#

I was about to bring up api_password, when I saw the space

balmy elk
#

you guys are the best. its working

#

shame on me, it was syntactically correct and everything. i just didn't see it

#

thanks so very much.

dense spadeBOT
balmy elk
#

Last Thing: I think its best to reopen the range for the proxy addon /container service from 172.30.33.9 to 172.30.33.0/24 cause container ip might change randomly from time to time in the given range (refer first post ...8 -> now ...9). This might weak security on one hand but otherwise there is a good chance for surprise with missing proxy accessibility, one day. A compromise here. Or do you have a better suggestion/solution here?

raven tinsel
#

I would advise against it

#

if you would still do that, sure, that is your call

#

nobody is stopping you

pine kestrel
#

hi, installed AdGuard Home and rebooted. However, after clicking on the menu item for Adguard Home (or clicking on 'Open Web UI') i get a blank screen.

#

ssl disabled, dns to 8.8.8.8,8.8.4.4

#

adguard Log stops at:
[15:35:28] INFO: Starting NGinx...
is that correct?

dim oak
#

which is why i had it too

#

@balmy elk what's your final snippet you used? i'd like to copy so i'm doing it correctly too

sly kindle
#

yeah old blog post vs docs

dim oak
balmy elk
#

like i said, not touched thos for years, and it even was not working because of that space 🙂

# 
# Uncomment this if you are using SSL/TLS, running in Docker container, etc.
http:
  #base_url: ******
  server_port: 8123
  cors_allowed_origins:
    - https://****my.super.io
    - http://****my.super.io
  use_x_forwarded_for: true
  trusted_proxies:
    #- 172.30.33.9
    - 172.30.33.0/24
  ip_ban_enabled: true
  login_attempts_threshold: 5
#

i changed back to 0/24 because of

#

A request from a reverse proxy was received from 172.30.33.8, but your HTTP integration is not set-up for reverse proxies

#

A request from a reverse proxy was received from 172.30.33.9, but your HTTP integration is not set-up for reverse proxies

dim oak
balmy elk
#

but after all, without the "space" config really works out much better, at all , hehe

dim oak
#

@balmy elk sorry for the dumb q: for cors_allowed_origins these are just your dynamic dns, right? so everything still works locally behind your firewall if accessing via local ip?

balmy elk
#

right. st the end maybe i dont need it in the confug, i just found out that my config here never worked here 🙂

dim oak
#

i appreciate the timing on your asking this anyway as my duckdns.org isn't working this morning 🙂

balmy elk
#

im my own dyndns provider :). cause i wanted a nice "own" domain here. But the situation after upgrading HA should be similar to yours, without knowing further details, i think

#

Well in the lasu 2 updates there were breaking changes with the proxy that you might use. Did you do an upgrade this morning? 🙂

dim oak
#

last night! ha

#

thank you for that info, i had no idea either

balmy elk
#

PS: Super Release-documentation. Really! If only my config would have worked 🙂

#

But my Blindness is not the responsibility of HA Team 🙂

dim oak
#

letsencrypt is working fine locally (192. etc).

#

i am wholly unable to access it via external dynamic url and not experienced enough with this to debug why

atomic drift
#

are you using the nginx add-ons ?

#

im using roughly the same configuration, im just not using duckdns but noip instead

#

letsencrypt working fine locally ??

#

it need a domain name to work properly

dim oak
#

thanks @atomic drift. no nginx add-ons, do i need to? and i was using noip as well but switched to duck because i couldn't get that working either 🙂 so i thought maybe it was a noip thing.

atomic drift
#

i think you need to use the nginx add-ons, and in your router you don't need to map 8123 but you map external:443 to 192.x.x.x:443

dim oak
atomic drift
#

id say yes you need to have nginx as reverse proxy, so nginx use the certificate from LE

#

at least that's my config

#

and it's working fine

dim oak
#

thanks, i'll do that now

atomic drift
#

my router config : ha_https TCP Port 443 192.168.1.xx 443

#

my nginx configuration

domain: YOUR_DNS
certfile: fullchain.pem
keyfile: privkey.pem
hsts: max-age=31536000; includeSubDomains
cloudflare: false
customize:
  active: false
  default: nginx_proxy_default*.conf
  servers: nginx_proxy/*.conf
dim oak
#

oh thx for sharing!

atomic drift
#

it may not work as is for your setup, but it should help 🙂

dim oak
#

i'm sure it will...thank you

atomic drift
#

yw

dim oak
#

for nginx, do we all always use trusted_proxies: - 172.30.33.0/24 by default or should i be changing that to something else?

#

i know zero about ngnix other than what it is

#

ok i just hosed something. i turned on nginx and now it doesn't work locally 🙂

This site can’t provide a secure connection

192.168.1.222 sent an invalid response.
ERR_SSL_PROTOCOL_ERROR
atomic drift
#

have you configured it to use the certs ?

atomic drift
dim oak
# 
http:
  # ssl_certificate: /ssl/fullchain.pem
  # ssl_key: /ssl/privkey.pem
  cors_allowed_origins:
    - https://xxxxxx.duckdns.org
    - http://xxxxxx.duckdns.org
  ip_ban_enabled: true
  login_attempts_threshold: 5
  use_x_forwarded_for: true
  trusted_proxies:
    - 172.30.33.0/24
#

trying to find my ngnix config via ssh...

atomic drift
#

oh

#

supervisor tab

#

then nginx add-ons, then configuration tab

dim oak
#

i can't get in 🙂

atomic drift
#

how so ?

dim oak
#

i have to do it via SSH now because my localhost has locked me out of web GUI

atomic drift
#

oh right

dim oak
atomic drift
#

running supervised ?

#

on hassos ?

#

or unsupported OS ?

dim oak
#

hassOS image on a rpi4

atomic drift
#

im running a different setup, so im trying to remember how things works on hassos

#

when you connect via ssh

dim oak
#

thanks! do you recall how i can disable an add-on via CLI?

atomic drift
#

yes gimme me 2 sec

sly kindle
#

@dim oak you commented out the ssl

atomic drift
#

ha addons stop <addons_name> may work

sly kindle
#

instead of https://

dim oak
dim oak
sly kindle
#

yes, which is true. but can you reach it using http:// and not https

dim oak
#

lol

sly kindle
#

there you go lol

#

when you use nginx you don't need to use https for local access anymore

dim oak
#

thank you for helping me find my ass with both hands hahaha

#

ok that's great. thank you!

#

still no joy from duckdns.org but at least my minor catastrophe is over

balmy elk
#

@dim oak to clarify im using hassio supervised with docker an the nginx addon. i can reach my instanz from in my networt directly by local ip (usually something like 192.168..:8123)

#

and i could rreach it "by Domain" from outside my network with my.home.io (example) . Thats where the proxy came in

#

the domain part was broken ecause of the new proxy settings to be made,.,

#

but ic ould reach my instance every time locally with it's ip

dim oak
#

thanks. for whatever reason i am able to access mine again locally (thanks @sly kindle ) but still no love externally. here's my config:

# 
http:
  ssl_certificate: /ssl/fullchain.pem
  ssl_key: /ssl/privkey.pem
  cors_allowed_origins:
    - https://xxxxxx.duckdns.org
    - http://xxxxxx.duckdns.org
  ip_ban_enabled: true
  login_attempts_threshold: 5
  use_x_forwarded_for: true
  trusted_proxies:
    - 172.30.33.0/24
# 
homeassistant:
  external_url: https://xxxxxx.duckdns.org
  internal_url: http://192.168.1.222:8123
  packages: !include_dir_named packages
  time_zone: America/Los_Angeles
  auth_providers:
    - type: homeassistant
#

letsencrypt works just fine as does duckdns as does nginx

balmy elk
#

if you try to reach your instance with the duck domain - is there any message in your ha log?

viscid orchid
#

is your goal to have SSL between NGINX and your HA instance? That's not typical

dim oak
#

@viscid orchid thx for asking -- goal is simply to get duckdns working so i have a static external! nothing fancy

dim oak