#add-ons-archived
1 messages Β· Page 57 of 1
you can send it to me directly
That don't seem to be working wither
I got it. Had to open the whole chat window to do it
you have the wrong nginx
Oh. This was the tutorial I was folling as this was what I managed to find that would allow me to use google assistant to play music and such..
the add-on i mean is the nginx with the white N in an green hexagon as an logo
le sigh... And that is the addon I need in order to use HA to play music with google assistant?
it is the add-on you need that you have accses to your home assistant instanz from the web
OKay.. So maybe I should start by telling you the problem... Because I could already do that..
My router does not support nat loopback. As a result (from my understanding) when I try and play music on my google speakers, it doesn't work. In addition, when I type https://<myduckdns>.duckdns.org into a web browser connected to my home network I am unable to access it.
However, the way it is set up.. If I am outside my network, and type https://<myduckdns>.duckdns.org into a web browser all works fine with SSL.
I see the problem, it is so that the router cannot tell a dyndns service what its current external ip is
is it all about playing music on the google assistant?
It would be nice to have the HA app work when on my network as well. But yea, google assistant playing local files I have uploaded to HA is the biggest.. p.s. Using a custom local DNS isn't an option either..
why should the app not work in the network, it establishes a connection when the smartphone is in the same network as homeassistant.
to the google assistant, it is a chromecast device you can add it via the chrome cast integration
Because the app needs SSL.. There is no SSL for 192.168.2.77 And my duckdns doesn't resolve to 192.168.2.77 it resolves to my public IP.
i donΒ΄t have ssl aktive on my home assistant and my home assistant app works just fine
Well, mine doesn't, and when I googled it.. It said that was why..
do you use the official companion app?
From the google play store, yes.
i simply entered http://homeassistant:8123/ into the address entry in my app
What I get is "webpage not available. err_empty_responce
Myabe its because I have non-ssl turnd off?
are you in your wifi?
yep.
is it even on your network? does fing show anything?
that was with just "Homeassistant" with "homeassistant.local" i get connection refuled
you have to add the port
Now, I hava managed to determine (I am quite sure) that non-http is turned off..
sorry, non-https is turned off
have you tried multiple ports?
I don't know what multiple ports is
other ports: regular http
I have nothing against enabling http again.. I just don't know how..
can you access configuration.yaml
yes
what http settings are configured
ssl_certificate and ssl_key ONLY, just those 2
if you remove them http should work
But then https won't work...
Is that not correct?
And I need the SSL because I use the google assistant to control devices.
i think you might be able to remove them and configure a proxy to have https seperately
theoretically you only need it if you want to access your home assistant instance from outside the network
Yea, we have been workign on that for over an hour... Apparently I was using the wrong addon.. So because we have someone new in the mix, I will tell you what the problem is...
This all started because I want to use music, currently on my HA, and have it play though my google speakers.
As I have already written, you can control the google speakers via chromecast
@royal valve Well, unless I am doing someting wrong... When I tell it to play music. I get a chime, and then nothing happens
you donΒ΄t need any ssl for connecting your homeassistant app in the home network to your home assistant instance
I go to "overview" I find a speaker, and I click in the bottom right the little play icon. I then select the music to play. Google speaker chimes. Then nothing
do you mean your media library
The music library comes up after I click play in the lovelace "dashboard"
okay what are the sources
Files I have already uploaded to HA
the chime only says that your google speaker is now connected
and they don't work right?
That is correct. I get a chime, and then nothing.
what is the format of the music
what if you manually go to the media library and choose to play on your device
let's revist the past for a moment: #integrations-archived message
so tts is similar to media library?
@viscid orchid good idea, lets check that
Okay.. So.. That may be the issue...... lol. Let me restart and see what happens.
if you do not want to access your home assistant instance from the outside, first deactivate all ssl add-ons
@royal valve I do want to.. And in order to allow google assistant to control devices, from my understand it is manditory
ah ok that wasn't clear to me until now
so far you have only said that you want to play music and access it with the home assistant app in the network
Okay, so I checked, and changed to make sure both internal and external url are the same, and are my duckdns url..
Played music, and same thinsg happened. Chime, then nothing
what audio format you are using
@royal valve They are mp3
hmm
Yea.. Thats what I have been saying for a while...
Try to open the same audio file in your browser. The full URL, domain and everything.
Oh, if I just click play, it plays through the webbrowser just fine.. I don't know what URL I would use.
But @cunning wolf that sounds like a good troubleshooting step... How do I lookup that url?
What have you been trying to use all this time? π€
@cunning wolf I am sorry, I don't understand the question. I have been trying to get music, located on my HA machine, in the media folder, to play though my google speakers
Just wanted to touch base. Its been 5 min and I havent heard anything from anyone.... Does no one have any more ideas?
Oh no, a whole 5 minutes.
People (and other sentient or vaguely intelligent beings) aren't waiting here for your questions, ready to leap upon them the moment they appear. Have a little patience. Keep in mind that repeatedly posting to get attention is against the rules (flooding) and can get you muted or banned.
Almost everyone here is a volunteer. It's also the weekend and they have their own lives to enjoy.
I am not upset... I am just used to people quitely getting frustrated with me and walking away because of my disability in needing a lot of clerification.
Not that I am saying that is what happened.. Sorry if I seem to have implied it.
I would recommend that you take this slow and look at it in smaller pieces. Check the documentation on the media browser so you understand where the media actually is. Try to play that file in Chrome or something and see what happens.
I would guess that it doesn't work the way you think it should.
@cunning wolf Well, I would assume if I hit play, a box comes up that lists the music, and I select the file, that music would play on the device I selected.. I am all for just not makeing assuptions most fo the time, but I think thats a fair one to make.. As far as playing in chrome. When I go to media browser, and selct the file, it plays fine.
I can see why you'd think it is... but that's not a fair assumption. There's stuff going on behind the scenes with security that you don't normally have to understand.
I'm not asking you to play it in the media browser within HA. I'm asking you to find the URL for the mp3 and to attempt to play that mp3 in your browser directly without HA involved.
because that's what the GH device is going to do
Okay, so I would like some advice on that please... But before that, you did make me think of something else... These files are VERY big... I wonder if that was the issue, so I am trying a smaller file..
So size of file seems to make no difference..
@cunning wolf I as for help locating that url because.. I don't know what I would google,, "When I use home assistant to play music located in my media library,m that is also located on the HA computer, to a google speaker, what URL does the google speaker go to in order to play the music" will not be very helpful.
For the love of dog, please stop tagging me on every reply.
HA has documentation. There's a page on the media browser functionality which will explain. Which is why I said... #add-ons-archived message
I don't have all the answers. Even if I did, just giving you all the steps won't help you learn. I'm trying to point you in the right direction so you can learn.
Actually, with my disability.. Providing steps is often the only way I learn.. I have a difficult time following written guidelines because assumptions that people naturally make, I just can't. And I understand that frustrates some people.
@worthy yew tagging people with every reply here in discord during an ongoing conversation tends to feel like using ALL CAPS because of the emphasizing and possible notification sounds
However.. I do think I see the problem.. Its been the problem I thoughtit was all along.. It loosk like the media has to be accessable via HTTPS... and its not.
If you have someone's attention, you don't need to use it
And, I don't know how to make it accessable via HTTPS on my local network. I can not use nat loopback, or custom DNS servers.
https://192.168.2.77 gives certificat errors (Of couse) and https://<myduckdns>.duckdns.org won't resolve inside my network.
Yea, I just can't find the URL the media files should be accessable to.. If thats the right words
By default, the integration looks for media in a specified folder. If other media_dirs are not declared you need to use /media/local path for example in companion app notification.
I don't understand what that means
So you're going to have a URL that looks something like: https://shadowwizard.com/media/local/name_of_a_song.mp3
Yea, I think I managed to track that down... Its using the external domain name.. That won't resolve internally.. I wanna see if the files are actually accessable as well..
So swap everything before /media with your IP address and paste that thing into a browser.
But... it won't work...
Files served from media are protected by Home Assistant authentication unlike those served from www.
No, i looked in the log. Its trying to cast from the external host name.
And agin, what you just told me, I don't understand.
I know where you're trying from now. I'm asking you to try the internal IP address instead of the domain...
What's the IP address of the device you run HA on?
Internal. Don't share your external IP here.
well, from an external domain, I get unathorised.. When connecting from outside the nwtwork
I don't care about your external domain.
Internally I get not found
192.168.2.77
And what's the name of the file you're trying to play? You should know what it's called because you put it into the media folder yourself.
Go to this address in Chrome: http://192.168.2.77/media/local/btb.mp3. What happens?
Connection refused.. But thats because for some reason, NOTHING works without https.
No, connection refuled again... Did we forget the port number?
8123
okay, will do again with poirt
with https: Unauthorised.
With http, "ERR_EMPTY_RESPONCE"
There we go. Ignore the domain. Ignore the fact that your router doesn't handle NAT loopback. It's because stuff served out of the media folder needs auth.
Do you have a www folder in your config?
Like in configuration.yaml ?
Not in that file. In the same folder as that file.
I don't think so.. But I can't see/understand the directory structor too well with what I have installed.
Create a folder called www alongside configuration.yaml. Copy your mp3 into that folder. Restart HA.
okay.. I have a terminal installed... Let me try that.
Looks like everything is in the config foler.. I assume thats correct?
Yes
Sorry, chould be more clear, looks like configuration.yaml is in the config folder.. So I will eb creating a www folder under config and putting the mp3 in config/www
cd ..
wrong window
HA is restarting
I now have btb.mp3 in www
However... Unless I have the URL wrong.. It is still not accessable.
Fixed
Well, its playing! With the https π
Great. So this is what I was trying to say earlier. /media needs auth. /www doesn't.
Okay...
You could use that URL that tom posted in a script/automation and your device should play it.
You can't use the media browser to get your device to play it.
Okay... And it will work, even with an invalid certificate?
Maybe. Try it.
But the reason I wanted you to check all this is because the domain/certificates definitely weren't your only problem. Start with the basics, then expand.
If you still need to sort out certificates, tackle that next. But it's late here, so I'll let someone else help with that.
I need to try and make an automation to do it... lol..
I think this is where it all goes pear-shaped. When I want to play a file to my Chromecast Audio I have to use a publicly accessible URL.
Tom, you know that still would not be the end of the world for me now that I think about it. It would be nice to be able to use an internal ip.. but I can deal if its not.
Internal IP here without SSL π Works great.
I know others do too.
Just gotta terminate the SSL at a reverse proxy like a sane person.
that was....uhhh... 3 hours ago
Is there no way to make HA reply both with SSL and non-ssl?
no
that's not HA, though
So, lets get one thing at a time going.. I am failing at creating anytying to broadcast to this device.. I must be googling wrong.. I use node0red a lot, if that helps, but at this point I just wanna see what works... How can I say "PLay <url> on <device>"
Keep it simple. Forget voice activation. Write a script... you can get help on that in #automations-archived
Thats what I am trying to do.... OKay, I will head over there...
Thank you very much for your help mono.. I finally managed to get it to play, at least from an external URL. I knwo it can be very frustrating trying to help me, and I am very appreciative of your paitence and workign with me.
I will come back somplace later to work on terminating ssl at reverse proxy, or whatever, lol
Anyone using HACS know how to see its installation progress or if it failed? I would suppose that when it's complete, the HACS menu item will how, yes?
HACS has nothing to do with addons. #integrations-archived was the right place
can anyone help me with cli commands to disable add-ons? I've had a failed update and want to try and reduce possible issues.
After "ha addons", I just keep getting list of addons ? or help brings further list
Hi there, use anybody Bookstack here? I just installed, but interestingly even though turned WYSIWYG on there is neither possibility to format the text nor chance insert a picture.
I run HA in supervised mode on RPi4.
Thanks for your advices.
Hey guys... I'm having issues with mosquito addon and would need to downgrade it to earlier version. I know this is not supported, but what are my options?
I dont have backups anymore, but I'm at home with linux / docker
I tried pulling older image, but it's not available on dockerhub anymore. Are containers even published there or are they built on install?
What kind of problems?
slow connections to the MQTT from my PC
ERROR: Can't setup Home Assistant service mqtt in mqtt log file
No ACK from MQTT server in 10 seconds in HA log files
my automations stopped working or they are slow to react
lights sometimes need up to 2s to turn on
and looking at https://github.com/home-assistant/addons/issues/1887, it looks like I'm not the only one
I have a question related to security. I want to set up influxdb and I am unsure about the ssl option. I just want to use influx in combination with grafana to display some graphs on lovelace. Does setting ssl to false lead to any insecure interfaces? I don't want public access to the database.
Anyone uses git integration in Visual Studio addon?
Is it possible to configure to show /config/.storage/* files in it?
@tidal tangle did you try un-istalling the addon and re-installing to start a new mosquitto.db?
Thit that now... We shall see:)
I look at the changelog of 5.1.1 and 5.1 and the changelog is only changing values of schema, so I have no idea why that would break anything... Error ERROR: Can't setup Home Assistant service mqtt also makes sense, since in run.sh, the script is calling HA MQTT service with connect data and since mosquitto is not running at that time yet (it it started a few lines after), it makes sense for it to error out.
not sure why this is even here as it should never return non-error response
and mqtt server is started at https://github.com/home-assistant/addons/blob/master/mosquitto/data/run.sh#L157
ApexCharts panel is not enough for your graphs? https://github.com/RomRider/apexcharts-card
settings > files: exclude
Does the auto-update option for addons also perform a snapshot as part of the process as a rule, or is that per-addon and dependent upon how it's managed?
after some more reading, it's there to register mqtt to supervisor. But I still don't know why it fails
I have trouble installing an addon. I click install but it's shows the revolving symbol for a few seconds and then goes back to showing "Install"
Any ideas where the problem lies? My setup or the repository?
go to the System tab and see what's in the Supervisor logs
Uh.. A lot of stuff related to the addon. Jees
Ending with sed: can't read forked-daapd.conf: No such file or directory
Removing intermediate container b7538ed4d04e
Trying another addon from the same repository:
Branch 'master' set up to track remote branch 'master' from 'origin'.
error: patch failed: Cargo.lock:1045
error: Cargo.lock: patch does not apply
error: patch failed: playback/Cargo.toml:17
error: playback/Cargo.toml: patch does not apply
Removing intermediate container 9b0a057b605e
Hi, I'm struggeling to get my hm-mod-rpi-pcb running with the ccu add on on a rpi 4 with a supervised installation of home assistant on rasbian lite, on start. I'm not even sure the ccu add on is recognising the device
Hey, I'm a little confused here, hopefully it's not a hard question... when I install HASSOS on my pi I can install addons through the UI, but when I use raspbian and docker image I can't - even though HASSOS also runs docker - what am I missing? Thanks!
the supervisor is missing in #449717345808547842
hi all
I recently installed wireguard and adguard on home assistant but since then my youtube has been forced to be in restricted mode.. can someone tell me where do i remove this?
what is restricted mode π
yea i had the same reaction at first
this is a pure youtube thing it seems, nothing to do with wireguard or adguard
no it is part of parenting mode from adguard i believe
yeah then change your dns server
it is preventing me to see comments on youtube
and i am unable to edit in settings because of adguard i think
rednode is #node-red-archived
Hi! I have a (simple?) question: I reverse proxy from an nginx to my Home Assistant, so all of the https and hostnames are on nginx, not HA. From HA, for all my add-ons that have a web interface, such as UniFi and TasmoAdmin, the "OPEN WEB UI" link is broken, as it points to my nginx hostname and the HA tcp port of the service. I've searched and didn't find a way to configure an override link for each of these. For instance I have the UniFi Controller reverse proxied from nginx on https://unifi.<mydomain>.com/ and could set that. Is it possible?
this makes no sense
you should get the pages delivered by ingress from your supervisor on the ha:8123 page
What do you mean? π€
The "OPEN WEB UI" links is currently https://domo26.example.com:8443/ because I have homeassistant: external_url: https://domo26.example.com
you should not need that external_url
But that's actually my nginx answering, and it's only reverse proxying 443/tcp to http://192.168.18.x:8123
Hmm, ok, but how will removing it help me with what I'm trying to achieve?
you are trying to access the addons page via the supervisor open web ui link?
Having an issue after updating to supervisor 2021.04.3. After upgrading it seems that i cannot access the addon section correctly when using my external URL. Internal works perfectly fine, but when accessing via the external and clicking on any addon, i get the error "Error while loading page addon."
the supervisor then grabs the page from the addon internal and returns it to you via the normal HA page. kind of like a iframe but very much different at the same time
@covert arch : Yeah, the link from the add-on's page is basically my external_url + the add-on's port appended (8443 in the case of UniFi Controller). I want to override that external_url for the add-on, since I have my nginx web server reverse proxying a different hostname on 443/tcp for it.
I've removed the yaml set external_url that I had set years ago, now I can change a bit more configuration from the web interface (which is a good thing π ), but the add-on's link is still the same.
I was hoping for a way to override the add-on's "webui": "https://[HOST]:[PORT:8443]" configuration somehow. I tried as yaml in the add-on's config (long shot), but it didn't work.
you donβt need to
this :8443 page gets loaded by the supervisor and presented to you on :443
no idea why you get these redirects
When I'm on the https://domo26.example.net/hassio/addon/a0d7b954_unifi/info page (UniFi accon page), and hover the "OPEN WEB UI" link, it points to https://domo26.example.net:8443/ ... so when I click it doesn't work, as the 8443/tcp port isn't open on my nginx.
Maybe I haven't been clear: This nginx web server of mine is a completely separate server, it's not an nginx add-on to HA.
To add to this, i would like to try and roll back my supervisor, but it seems that the syntax provided was not working. I tried ha supervisor upgrade --version 2021.04.0 but i got that the error of "no supervisor update available"
this should not happen.
it should direct to https://domo26.example.net/hassio/addon/a0d7b954_unifi/something
but vip3rr pretty much reports the same issue, maybe there is something going on
For me, it's been like this forever. This is a HA that I've been updating for a long time... it's probably full of legacy stuff and leftovers. Maybe that could explain why this "magic" handling of add-on web interfaces isn't enabled π€
Mine works completely, its JUST the addon page. I can access everything via my reverse proxy external URL, this just broke when i updated the supervisor, and now i cant seem to downgrade it
@lean garnet what's in the supervisor logs?
@severe sandal Is it only the unifi addon? It doesn't have ingress support.
sadly nothing indicating that there is even an issue at all
@sly kindle : UniFi Controller and TasmoAdmin. Both have links to my external_url + their custom http port.
@lean garnet ctrl + shift + r a few times while on the supervisor page?
I did see this for the UniFi addon:- This add-on cannot support Ingress due to technical limitations of the UniFi software. but I had no idea what "Ingress" actually meant in HA terms π
Now it's a bit clearer... but I'd still like to be able to set something like an external_url just for that add-on, since I have my own reverse proxying configured to it on a dedicated hostname.
@severe sandal yeah neither of those will open behind HA, which is what ingress does
So those need to be configured manually for external access.
Yeah still nothing in the logs indicating that there is even an error occuring. Is there a way to downgrade the supervisor? it was perfect until this today
I am now on supervisor-2021.04.3
same here. no issue with it loading the page
You using a reverse proxy too?
I am not. Does it work when you are using local?
Yes it does, but again this behavior did not start until i upgraded the supervisor this morning
@sly kindle : Understood. But can I override the (broken) link that HA shows from "OPEN WEB UI" for those? Because it's just appending their ports to my external_url, which doesn't work.
@severe sandal can you reach them when manually entering the correct url?
Yes, I've configured my own nginx for them. I have https://unifi.example.net/ going to HA:8443 and working fine.
Using reverse proxy here (NGinx based), not experiencing the issue described
Yeah and i have nginx reverse proxy manager.. maybe i should try it differently, make no sense that it all the sudden broke for me
FWIW, I was running 2021.04.0 until a few minutes ago, and now 2021.04.3 still works fine for me too π
@severe sandal there's no way to change that open web ui link, so just ignore it. You just "know" what the URL is and bookmark it or whatever
Make your own link in the side panel or UI
Yeah, but every once in a while I go to HA and update my add-ons, then I want to check they're working, and just click away... and always get annoying for not being able to reach those trivially.
I know, I'm picky, sorry! π
And the "webui": "https://[HOST]:[PORT:8443]", line of config.json is mocking me! π
...and replacing it, then rebuilding the container has actually crossed my mind. I better move onto something else before I do something useless and ugly that I'll regret! π€£ Thanks for the help!
I just installed AdGuard Home (through the Home Assistant OS), and updated my DNS server (in router settings); however, I'm not getting any queries or blocking at all. Any thoughts on what to check?
@quick flume you may need to reconnect some devices to the network?
What OS is used on the device you are browsing with?
Host = Home Assistant OS β> AdGuard Home
Client = Windows 10 PC
I've configured the router settings to use AdGuard Home as the DNS server, and devices are able to browse, but no queries are being logged in AdGuard Home.
have you confirmed the windows pc is actively using the new dns server?
I have to enable/disable ethernet to get immediate effects when I change dns
ipconfig /all in cmd on the windows pc would show you
Yep, it shows the DNS server that I'm expecting as well. I'm wondering if it's just an issue with AGH in some way
you also want to set DNS servers on the HA Host
Hmm, how would I do that?
supervisor > system > host card
click on change across from your ip address
set ipv4 to static, and enter dns server that isn't adguard. google's or cloudflare or quad9 are what I normally go with
you don't want your HA host have to go through itself to get dns, because after a reboot, that will end badly
adguard will not be running
Ooof, okay β I'm going to change DNS Servers there now. I guess the best option is to just use what AGH will be using anyways
So https://dns10.quad9.net/dns-query in this instance
I think those are cloudflare 1.1.1.1 or quad9 9.9.9.9 by default
this is also why I eventually stopped using the addon and started running adguard elsewhere
Am I the only one with problems with MQTT 5.1.1 . New installation. Like this https://community.home-assistant.io/t/mqtt-suddenly-not-working-2323-error-cant-setup-home-assistant-service-mqtt/288632. And more important; what to do??
Gotcha, so ETH0 is updated and now using 9.9.9.9 β I'll restart everything and see if it'll start logging queries
@velvet breach what does your addon config look like?
No change unfortunately =/
Full config is:
ETH0 > IPv4 > Static
IP address/Netmask = [ha host]
Gateway address = [router]
DNS Servers = 9.9.9.9
logins: []
anonymous: false
customize:
active: false
folder: mosquitto
certfile: fullchain.pem
keyfile: privkey.pem
require_certificate: false
@velvet breach ok and this is a fresh install of the add-on? What is your hardware setup?
Hi guys, from tonight i got a problem with my Network UPS Tool, it give me my devise unavaible. Addon log say this: https://pastebin.com/d2nk43eh
I tried to update HA but it continue to doesn't work.
By the way it is about 2 week I don't change/update something on my HA, so don't know what can be.
Now i'm looking better, and i see conbee 2 stopped to work at the same hour of NUT. Can be a USB problem? My HA is on rpi4 with SSD via USB
Fresh install, yes. ESXi with Home Assistant, and the client trying to push some MQTT-data to it, is a RPi Zero.
I'm at a bit of a loss unfortunately. Checking logs, the only things that stand out are:
2021/05/03 12:27:40 [error] Couldn't load filter 1 contents due to stat /data/adguard/data/filters/1.txt: no such file or directory
2021/05/03 12:27:50 [info] Couldn't request filter from URL https://adguardteam.github.io/AdGuardSDNSFilter/Filters/filter.txt, skipping: Get "https://adguardteam.github.io/AdGuardSDNSFilter/Filters/filter.txt": couldn't initialize HTTP client or transport, cause: couldn't initialize HTTP transport, cause: couldn't bootstrap https://dns10.quad9.net:443/dns-query, cause: failed to lookup dns10.quad9.net, cause: read udp 192.168.1.229:49388->1.1.1.1:53: i/o timeout
2021/05/03 12:27:50 [error] os.Chtimes(): chtimes /data/adguard/data/filters/1.txt: no such file or directory
2021/05/03 12:27:50 [info] Failed to update filter https://adguardteam.github.io/AdGuardSDNSFilter/Filters/filter.txt: Get "https://adguardteam.github.io/AdGuardSDNSFilter/Filters/filter.txt": couldn't initialize HTTP client or transport, cause: couldn't initialize HTTP transport, cause: couldn't bootstrap https://dns10.quad9.net:443/dns-query, cause: failed to lookup dns10.quad9.net, cause: read udp 192.168.1.229:49388->1.1.1.1:53: i/o timeout
Completely brand new fresh install now, still having the same issue on both addons. I can get to the addon page and see the addons, but there are no labels are at the top for Dashboard, add-on store, snapshots, and system. when clicking on one of the addons, i get that "error while loading page addon." error
Not really sure what to do since this was just working completely fine before
Maybe it isn't the system, but the browser?
Interesting thanks for the tip. Suprised Chrome is so messed up, Brave is working i really appreciate it
Chrome 90 has some weird issues: #330990055533576204 message
chrome 90.0.4430.93 here
Actually i take that all back. I had http:// for my external url and it was working. Put back in https:// and now its doing the same thing again
for troubleshooting i put the http:// in chrome just to see and it popped up
so seems to be an SSL issue
Proxy is definitely working and reverse proxied to Cloudflare, but its almost like not all of the traffic is going there. Really weird that thats not rendering under https
Thinking about just moving the supervised install to docker on my rpi4. Have the rpi4 with 8gigs of ram, might not be a problem to run it without the supervisor
I did have to add a separate location for websockets, but I think it was just for some of the "open web UI" links, such as for ESPHome. I use a location of /api/websocket redirected to my HA server and the following additional config:
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
but I'm just a script kiddie when it comes to this stuff
Any idea?
@sly kindle for reasons I can't explain, it's now working.....
Oh snap... I think I got it
My upstream DNS server was set to https://dns10.quad9.net/dns-query
I changed it to 9.9.9.9
Found it in https://community.home-assistant.io/t/home-assistant-community-add-on-adguard-home/90684/509
Hi all, I have a problem with the Unify Controller add-on. It worked till last weekend, but now it won't connect anymore.
I see this in the logs:
Logger: homeassistant.components.hassio
Source: components/hassio/websocket_api.py:106
Integration: Home Assistant Supervisor (documentation, issues)
First occurred: 8:01:05 PM (3 occurrences)
Last logged: 8:10:05 PM
Failed to to call /addons/a0d7b954_unifi/stats -
Failed to to call /supervisor/stats -
Failed to to call /core/stats -
I get the: ERR_CONNECTION_REFUSED message. I run supervisor-2021.04.3 and it also says unreachable
Hmm.....it only worked for a short time
1620063325: New client connected from 10.0.0.240 as auto-77F7D804-D489-5FF3-9EA7-A4B4CD811DD1 (p2, c1, k60, u'mqtt').
1620065027: Saving in-memory database to /data/mosquitto.db.
1620066051: Socket error on client 0xqmXVFpYEU2AkXQXOJ7av, disconnecting.
@velvet breach is that client using the same authentication as others?
In the logging I only see: 21-05-03 18:37:49 ERROR (MainThread) [supervisor.utils.codenotary] Timeout while processing CodeNotary
And I've turned the system off and on again, which did help a bit, but not for the Unify Controller add-on
I've also read @lean garnet his problem, but I have no special setup. The only special thing about my setup is that there is a Ubiquity USG, a switch and Ubiquity AP are in the network (hence the Unify Controller add-on π )
Anyone knows if it exist a facebox and double-take addon to ha?
Hi Vip3rr, I had the same behavior ("error while loading page addon.") and after clearing chrome (ctr+shift+R) the page was loaded correctly.
I get now this in the logging:
2021-05-03 20:48:26,833 main ERROR Unable to locate appender "InMemoryAppender" for logger config "root"
(I finally got logging......)
Hey all thanks for all the responses. I will keep working on it and will hopefully have a resolution
What's also strange, according to HA the controller is running, but if I try to connect to it through ubiquity cloud, It states it's offline.
I believe so. Can you tell me what to check?
It seems that https://community.ui.com/questions/Error-on-startup-of-Unifi-Controller-on-Raspberry-Pi-Error-processing-element-InMemoryAppender/4ed28ed2-fa98-4d7a-b242-80a4100b7529#answer/80bfee48-49a0-48e5-9c9c-090b112a51fa is a fix for it (or an update to version: 6.2.17)
But I don't know how that fix can be done. @lean garnet maybe that is a solution for your problem as well?
I may have had a caching issue like people are advising above. Seems like its working ok right now, will try and setup the rest of my HA and will report back when i have a chance to complete
I tried the CTRL+Shift+R used the incognito browser, used edge (which I never use), but still the same error.
I don't know if anyone sees my messages, but I hope you do π
Hiya, I'm trying to get Rhasspy to output sound on my Raspberrys 3.5mm. The aplay commands Rhasspy uses to play the generated wav does not work inside Rhasspys docker, but works just fine in a hassio terminal. Inside the Rhasspy container I get an error from aplay about Unable to install hw params. Any ideas where to look? Smells like permissions?
Hi @vapid token , I see that there is also a #voice-assistants-archived channel, maybe that your question can be better answered over there? Just suggesting, I'm totally new here
Ah I didn't see that @vestal kestrel, good one! Thanks
Is this where I can ask for help with node-red integration? I've had it working fine for a while, but for some reason now the server-events nodes are all stuck in 'running' and wont trigger the flow
thank you
Installed SSH & WebTerminal and got it "working", issue being resolution in Web terminal is weird (201x52), not readable at all. Can someone please help diagnosing?
@stark grove what browser?
I have tried Brave and Firefox. The latter could not open the WebTerminal at all, just said "Connection Closed" when clicking it
Hi i would like to save the DNSMASQ addon log on a location i can consult with node red, is this possible ? i cant find a clear solution online
Does anyone know why the different bitwarden clients requires different configuration on where to find the bitwarden_rs server? Im running the bitwarden_rs addon in my Homeassistant Blue but the different clients seem to have different ways of needing the serverurl provided. The ios and windows-app for example, needs the server ip and port to be configured in "Custom environment - Web Vault Server URL", but in the firefox-extension, the server ip and port can be configured in "Self-hosted Environment - Server URL". If configured otherwise, both 'groups' of clients throws "failed to fetch"- errors.
What is even more wierd is that if i enroll new users via eg. the windows-client with the current configuration. That new user will not be displayed in Bitwardens sys-admin portal, and its not possible to log in as the new user via the web-vault(wrong usn/pw error). Although no errors seemed to arise in the enrollment-process. Also when logged in to the windows-client with the new user, syncing and such causes no errors in the gui.
Hi all, is it possible to install this add on with a Virtual Venv installation? https://github.com/korylprince/hassio-apcupsd/tree/master/apcupsd
@proven tide there are no add-ons with a venv installation
that sucks π¦
and that particular one has been archived
gotcha
add-ons are docker containers. You'd have to be running the ecosystem (which uses docker) to have the add-ons
i see, makes sense
@proven tide you can still use the integration https://www.home-assistant.io/integrations/apcupsd/
How so? It says I need to install that addon?
with the install somewhere else
not sure if i follow, i don't use docker so how would I go about installing that add on?
the add-on would provide the service if you needed it, or didn't already have it installed somewhere else (some windows pc, os x) http://www.apcupsd.org/
the integration has been available long before the HA OS install existed
oh, so if I just install apcupsd with apt-get I can then add it in the config.yaml?
right
#551864459891703809 for your install method. and #integrations-archived for help configuring the integration itself
thanks!
Does anyone know how to SSH into an add-in - specifically, the Unifi Controller Add-In? I'm not sure how the add-ons are done. I would think that they might be docker containers. I need to fix something that can only be done by SSH'ing into it. I doubt that SSH'ing into Home Assistant will do it...
They are containers, but you donβt ssh into containers
You start a shell in them with Portainer or docker exec
But you really shouldnβt be doing that
I just need to update something in the underlying mongo database... Unfortunately, I can't do it from the UI. It's a known issue with the current Ubiquiti controller software.
@viscid orchid So I have Portainer installed on my HA instance. I do not see any "containers" - but I do see a bunch of "images" - that seem to correspond to my add-ins.
They all say "unused".
Theyβre filtered. You can delete the filter
I see 24 containers on the dashboard. I can't, for the life of me figure out where this filter is that is hiding them...
(not the dashboard... the "home" screen)
Didnβt look in settings?
I've looked all over.... I don't suck this bad....
Oh.... wait... I see 10 "hidden" containers - with values such as "base", "core", "supervisor", and "addon" (in addition to others). There is a "remove" button next to each one. I'm hoping that "remove" means remove from being hidden - not remove the container itself.....
@magic mist no no no
OK... @sly kindle you've got my attention....
don't remove your containers
I really hope I don't.... LOL.... I just want to get into the Unifi addon container just briefly. I can do a backup of HA, as well as the Unifi config. How can I get in there with a command line?
I don't use portainer but it's in settings, you remove the hidden label or something
Yeah, I see that. And I've been testing it out with a test container that I spun up.... I can add and remove name and value pairs all day from the "hidden" containers. But I have yet to successfully hide my test container.
I'm not sure what those name/value pairs refer to..... Maybe some sort of "label"...? but I can't seem to add a label to my test container. All these default hidden containers have a "io.hass.type" name, and (for example) "addon" as a value.
testing labels
Donβt try to add a label, delete the filters that are hiding the other containers
I was just doing a full test of the "remove" function. I added a label to my test container and I was able to add/remove name/value pairs and show and hide that test container, based on my label.
So now I feel confident that if I remove the "io.hass.type"/"addon" name value pair - it won't remove any containers, and I can restore it back the way it was, after I am done.
Is anyone here using frigate?
um, yes
Hello, need help with tasmota add-on
Switch is working but after 9.4.0 update I didn't get additional sensors like ip or ssid
Hello, i've added the zigbee2mqtt repository URL to the addons store, but even after restarting, it is not showing up in the store
and I don't understand why
what url did you add?
the one the readme said ^^
https://github.com/zigbee2mqtt/hassio-zigbee2mqtt
it shows up in the list of repositories with Home Assistant Community Add-ons
hm yea that is right. So after refreshing you don't see that repository in the list of repositories in the add-on store?
exactly, even after restarting the whole VM
you should be able to go to Supervisor -> Add-on Store and then see a repo with its own addons, probably at the bottom since it starts with Z
eeeeer, it shows in the lsit of repositories osrry, the addons themselves doesn't shows up in the store
http://prntscr.com/12j7l72 both officials and communiity addons shows up, but not the zigbee2mqtt
you said you refreshed the VM, did you refresh the UI?
The UI has aggressive caching, new repos don't show up for me until I refresh my browser
there we go. Yea you usually don't have to restart the system, just refresh the screen
Hey hoping someone can help me..
I have several add-ons but they are not auto starting when I try to start them I get a docker error.
When I update them I get a error saying the system is unhealthy.
Anyone?
share dockers error log
How can I locate that to share
I am new to Hass though
docker log homeassistant
or docker log --follow homeassistant
From the console
yes
user name or password for ?
The console
21-05-06 11:58:16 INFO (MainThread) [supervisor.snapshots] Found 6 snapshot files 21-05-06 11:58:58 WARNING (MainThread) [supervisor.jobs] 'AddonManager.update' blocked from execution, system is not healthy 21-05-06 11:59:52 WARNING (MainThread) [supervisor.jobs] 'AddonManager.update' blocked from execution, system is not healthy
I think it's because my docker version is 18.09.7 and it's running on Ubuntu 18.10
Hi I wonder if someone could advise me on this please. I ran check home assistant configuration and got this error. 08:47:58] ERROR: The configuration check did not pass!
[08:47:58] ERROR: See the output below for more details.
Testing configuration at /tmp/config
Fatal error while loading config: (idna 3.1 (/usr/local/lib/python3.8/site-packages), Requirement.parse('idna<3,>=2.5'), {'requests'})
Failed config
General Errors:
- (idna 3.1 (/usr/local/lib/python3.8/site-packages), Requirement.parse('idna<3,>=2.5'), {'requests'})
Im trying to add a tuya dimmer switch to localtuya but cant seem to configure it as a dimmer. Is there a configuration specific for dimmers
Hi can some one please tell me why i cant add !secret SHHuser in the configuration for the ssh & terminal please . it states invalid yalm
*yaml
Tuya adding the device automatically, do you check the entity in the dev view?
Or check what tuya sad in the Integration view
It doesnt the device automatically. I have to add the device manually in localtuya
Add
Having trouble with the Zwave JS add on not properly saving my network key
It always rolls back to 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, 0x10 despite setting it to a proper one like A9A1BC858ED640BEBE064E9211609582
Have tried both the UI and the yaml editor, doesn't make a difference
...and actually, /data/options.json in the docker container seems to have the correct key in it? but the UI does not...
very odd
@strange cove posted a code wall, it is moved here --> https://paste.ubuntu.com/p/sYsvjjdvrK/
Oh snap:D
To make it easier to read:
I don't know if this is the right channel to ask. I have a little problem with esphome.
I want to send a float value via http_request.post.
This is the code:
Sadly, when I try to compile this I get the following error:
config/esphome/esp32cam01.yaml:75:14: error: could not convert 'my_value' from 'esphome::template_::TemplateSensor*' to 'std::__cxx11::string {aka std::__cxx11::basic_string<char>}'
return id(my_value);
So I guess i have to to some kind of conversion. I googled but I did not find a solution for this. Maybe someon can give me a hint?
I hope this is okay?
Hello, my name is Elias. I'm from Germany and need help with the samba add on, I can't find my homeassistant in the Windows explorer? pls help me.
I got homebridge working with the ffmpeg with notifications to my apple tv as doorbell trigger via ftp upload, would love to replicate with home assistant. can anyone help me work this, doesn't seem to be an easy way
Opened a bug for my issue here, hopefully someone can help: https://github.com/home-assistant/core/issues/50172
Im using the Visual Studio Code plugin, and have the lovelace_gen HACS integration installed, which requires using "{{ variable }}" but when i save the file in VSCode, it reformats as "{ { variable } }". Does anyone know how i can stop VSCode from reformatting these brackets?
@clever rampart what do you mean?
I feel like I am missing something with this Chrony add on... I definitely get the date and time, but I can't figure out how to reroute all outgoing NTP traffic to the homeassistant to actually utilize chrony. Or how to tell it's working properly LOL
@dark gorge it's.meant to be a local ntp server for whatever you want to point at it, but can keep your HA OS host updated as well. There is no real home assistant integration for it
@sly kindle I just don't know how to tell if it's working... after that I was going to use my firewall to port forward all port 123 requests to it. but then i have to force it to use a different port to go out and update lol
i don't know if that makes sense... i figure it's using port 123 on the server for NTP requests... i just don't know how to configure it to use a different port for it's own NTP requests out
The port itβs listening on and the port it makes requests to are unrelated
It will use some unused outgoing port to connect to port 123 on the remote server
so how can i test that requests to it are working properly???
i know it sounds dumb, but i just want to see its active and working
This is weird. I updated File editor to latest, and suddenly the "Overview" shortcut points to "0" instead of "default_view"... How can I set this back?
Restarted Core and a few minutes later it had corrected itself... What can this have been?
Hi everyone! Can I use the VLC add-on to play TTS notifications locally?
When I use MQTT explorer to login to the Mosquito broker I'm running as an add-on, I don't see any topics at all. Just blank. It's working though, because Zigbee2MQTT is utilizing it
Anyone got Google Assistant to work with Sonarr/Radarr? I.e commands to add movies using assistant
hi, I'm thinking of trying out adguard home a new pi (not running home assistant). is there a way I can control this instance of adguard from homeassistant running on a different pi?
(I had this setup via creating a switch for pihole )
I just found out that the android ADB is discontinued on HA. But without it using the standard media integration i have allot less info getting out of the new google tv. im not sure if this was the best choice..
Anyone could help point me the right way? With the deconz addon turned on my HA instance becomes so unstable and drops out every few seconds. What can I do?
If I ping my HA instance from a local machine connected directly by ethernet it is stable until I start deconz, navigating the HA gui becomes super sluggish and my ping (to HA) times out every few seconds
Driving me crazy.
does the wireguard add-on support connecting to an existing server? i only want my hass system to be a client.
It is a server
Itβs a vpn that lets you connect to anything on your home network
As does any vpn
most non tech people in my area still think vpns are for torrents only
well after that whole ordeal of people getting sued for illegal movie download it kind of shoves that mindset forward lol
does anyone know how to forward ports to a portainer container?
Yesterday, I installed HA_OS on a rpi4 and having problems with SSL on the network. I am using duckDNS, MariaDB, Ngnix PM addons. I have a secure remote login- that works. I was trying to accomplish the same for lan as well. The documentation for the duckDNS indicates an additional need to add http configuration in the conf.yaml. When i add that bit of code, i can no longer access HA via browser. There is a lot of information found on how to do this but I am not sure what is current and what has been superseded. I have removed the http section from config.yaml so I can still access HA. I would appreciate help in getting letencrypt working on the LAN. TIA.
how do I keep the nest integration from insisting on a reauthentication every few days?
I want to authenticate once and have it just work
Hello everyone. Is there a way to enable an addon to show for a user that does not have admin privileges? I am specifically looking to try and use the Grocy addon for common tracking and shopping lists between my wife and I.
if you let them connect directly to a port instead of using ingress, yes
I cant seem to add cloudflare addon? everytime i enter my global api key?
what am i doing wrong
what error?
Invalid authentication
more context?
@cosmic bane I got it. It was setting mistake
I'm using the Z2M addon. I'm still on version 1.18.1-1. On the GitHub page of Z2M addon the newest release is 1.18.3-1. How can I check the reason of not seeing the new version ?
There's a button to reload in the three dots of the addon store tab.
If anyone runs into this, plug your conbee II to a USB 3.0 port instead of 2.0, solved it for me. 3.0 is supposed to have interference issues with Zigbee but it's finally running smoothly now that I switched to 3.0.
is there a way to run a script to add EcoPlugs intergration to my HA?
i cant edit the configuation.yaml nor add the files for the custom add on to the HA folder as it is marked Read Only
its a custom component
found this in the Communtity
https://community.home-assistant.io/t/eco-plug-workchoice-woods-wion-kab-etc-component/57297
if your HA directory is read only you have other issues
you should absolutely be able to write to configuration.yaml
I Have HA running as an snapd add-on in Ubuntu 20.04
Has anyone solved the issue of supervisor not being able to connect to Github?
[supervisor.store.repository] Can't remove built-in repositories!
Some mention here, but I'm not really seeing a proper solution:
https://community.home-assistant.io/t/no-valid-repository-for-https-github-com-home-assistant-addons/292011/77
@void timbertried an ha su repair ?
I've not, I'm running on a Home Assistant Blue, so trying not to get too involved in command line monkeying
I will do that, but I'm seeing enough other people having these errors, and them coming back, that I don't feel like it's just my installation.
Ran ha su repair and restarted supervisor, same thing
I've disabled DNS caching on my Eeros and all security settings there. There's something going on in the recent verions that won't let the supervisor connect to Github correctly (see the link above)
21-05-10 17:11:14 ERROR (MainThread) [supervisor.store.git] Can't clone https://github.com/home-assistant/addons repository: Cmd('git') failed due to: exit code(128)
cmdline: git clone -v --recursive --depth=1 --shallow-submodules --progress https://github.com/home-assistant/addons /data/addons/core.
If I try to manually add https://github.com/home-assistant/addons to my add-on repositories, it tells me that's an invalid one.
If I do nslookup github.com from Home Assistant, I get: ** server can't find github.com: SERVFAIL (nslookup works fine on other computers on my network)
@void timberwhat dns servers are showing at supervisor > system > host card when you click on "change" across from your ip address
I'm using DHCP, but if I switch to Static it shows Gateway 10.0.0.1 and DNS as 10.0.0.1
My eero's are using the default ISP DNS (75.75.75.75)
@void timber try changing dns to 8.8.8.8 or 9.9.9.9 and then save and reboot from the same host card
no dice
WARNING (MainThread) [supervisor.store.git] No valid repository for https://github.com/home-assistant/addons
It's something weird about the Supervisor though, as HACS can connect to Github and get repos fine
do you have a snapshot you can restore?
i have many, no idea which version this started happening with, but it's also not just me.
https://community.home-assistant.io/t/no-valid-repository-for-https-github-com-home-assistant-addons/292011/77
Also seems to be something that's been broken before:
https://github.com/home-assistant/supervisor/issues/2262
And I can ping 8.8.8.8 from the terminal, but nslookup for github.com is a SERVFAIL, so it's not internet connectivity
Is there a way to copy my database for grocy from one machine to another? I have rebuilt my home assistant on a second machine because I had a lot of accumulated issues that i didn't want to transfer with a straight backup.
trying to add the following lines to the dnsmasq addon, but cant work out the forma of the lines needed (under services?) thanks. dhcp-range=192.168.30.255,proxy
log-dhcp
enable-tftp
tftp-root=/srv/tftpboot
pxe-service=0,"Raspberry Pi Boot"
BTW, restoring a snapshot doesn't help, as the Supervisor version doesn't roll back with a restore, and it's Supervisor that can't connect to Github (HACS and everything else can though).
I worked it out, each line needs a colon after, and then any remainder hyphened below
Hi I wonder if someone could advise me on this please. I ran check home assistant configuration and got this error. 08:47:58] ERROR: The configuration check did not pass!
[08:47:58] ERROR: See the output below for more details.
Testing configuration at /tmp/config
Fatal error while loading config: (idna 3.1 (/usr/local/lib/python3.8/site-packages), Requirement.parse('idna<3,>=2.5'), {'requests'})
Failed config
General Errors:
- (idna 3.1 (/usr/local/lib/python3.8/site-packages), Requirement.parse('idna<3,>=2.5'), {'requests'})
Hi everyone I'm new so please bare with me π I encounter the problem with vlc telnet connection that no matter what host I provide in configuration.yaml the log allways shows [0000007faee79ec0] [telnet] lua interface: Listening on host "telnet://:4212". with no host included. What am I doing wrong?
I noticed this error: Error occurred loading configuration flow for integration vlc_telnet: No module named 'homeassistant.components.vlc_telnet.config_flow' it's normal add-on installation
Hi guys, can you recommend an add-on that updates the DNS record of my domain? I used Duckdns, but now I changed to a custom domain.
Hi guys, I'm getting a notification from my Home Assistant instance saying that the add-on samba_backup uses secrets which are detected as not secure. What should I do to secure it? If I put the login credential in the secrets.yaml the add-on configuration gives me an error.
the password you are using the actual password has been leaked in a global database before
@hollow imp It depends on your DNS provider. Thereβs an integration (not an addon) for Route53 (mine), and maybe thereβs one for yours
Hi people, I've a problem with NUT addon.
lan below the trace:
[18:22:24] INFO: Starting the UPS drivers...
0.000000 Starting UPS: upssotto
Supported UPS detected with megatec/old protocol
45.003783 Startup timer elapsed, continuing...
Network UPS Tools - UPS driver controller 2.7.4
[cont-init.d] nut.sh: exited 1.
[cont-finish.d] executing container finish scripts...
[cont-finish.d] 99-message.sh: executing...
Any ideas/suggestion?
looks like your UPS didn't respond in time
Thanks but so what i should do to solve this?
change your password
@viscid orchid .... Hum is new, has 3 month... I can restart it...
Is possible to debug the usb connection or tracing the error?
This is happened after a NUT upgrade.
your options are limited within the addon
I don't use the NUT addon, so I don't know where you would go from here
@viscid orchid thank you... I still harvesting info;-)
Hey all! I'm using the File Editor add-on. When I try to download an entire folder, eg Config, I get a 502 error. Is that expected? I would expect when downloading a folder that it'd zip it and give me that as a download
I initially thought it was an error on my DNS config, but got the same error when accessing via myIP:8123
What is the url for motioneye when Its hosted on haos itself?
[6:01 PM]
Do I need to send it to a specific port?
@bold dawn I would recommend using an actual file transfer protocol like Samba to access the filesystem, rather than the File Editor addon. I get the same thing, and it's possible that it's just not supported
it could also be that I'm going through NAT reflect and NGINX, too
Nope, I get the same "502" error when I connect directly. But it's not the right choice for file activities like that
looks like this is the error from the Supervisor: 21-05-11 22:26:07 ERROR (MainThread) [supervisor.api.ingress] Ingress error: 400, message='invalid constant string', url=URL('http://172.30.33.5:8099/api/download?filename=/config')
yeah, this or that file is going to be fine but a whole directory is not going to go well
use smb or ftp etc
how do !secret's work in add ons? I try to direct it to my secret.yaml but it says invalid...
Hey - did you get anywhere with this? Iβm trying to do the same but getting unknown error
I think it might be the container name but even with the admin login it errors out for me.
i'am trying to install and running FaserF Apache2 addon, so i have a mistake at start. i must uncheck SSL to run without error (i don't know about manage cert file) ; web file architecture is found (index.html) and i have a 502: Bad gateway when i click on sidebar link
@dreamy minnow when you enter it, use quotes or single quotes around the whole thing
username: '!secret sshuser'
password: '!secret sshpass'```
my addon Apache2 run but i'am using framework who needs to create file cache in/share/htdocs/tmp directory, how to put R/W right for Apache2 addon ?
@gilded totem I'm not familiar with that add-on but if it has access to /share then...it has access to /share?
addon must create some files in /tmp directory and i don't know how to put rights to solve "Access denied"π₯Ί
Hi all. Kinda new to hassio here... I setup hassio on a Pi 3B+ a while back. Have a working NodeMcu already running flawlessly. I remember that I've used the ESPHome addon to upload the firmware (1st time), and changes the YAML file from there. Although I see the device as ESPHome, I don't see the EAPHome addon at the add-on page anymore, nor at the Add-on store. I was looking for the 'flash' menu at the right hand top corner, if memory serves me well, but where did it go? System is using the latest versions of Supervisor and Core as of today. What I'm missing? Thanks in advance.
@wild furnace are you seeing the community add-on repo?
Nope, just the Official Add-ons. π€
that's where the esphome add-on is
Ok, added, and shows up. Thanks. So looks like the actual support is native now, but somehow I need to install the Add-on to get the ESPHome page. π
the add-on will provide that, yes
i have the mariadb add on but i need to delete the database and rebuild. i cant seem to find where tyo do that?
@ember crater uninstalling it and reinstalling would do
ok i was only restarting it and it wasnt enough. ill try thatr
is it just the one home assistant database?
This would be better: https://discord.gg/FYRnmtPm
Hey guys! how do i permanently map a directory to a addon container?
You donβt. What are you trying to do?
trying to map /var/run/docker.sock to node-red container so i can control a game server from node-red π
yeah, you don't
it works, not persistently though
i'll have to run node-red outside of HA then
big sad
is there any add-on to monitor the health of remote (and local) drives ? smart values are what i'm after i guess, would be quite neat π
hmm i guess i can simply dump the smart data to json and send that to HA via mosquitto_pub, yeah that will be a good way on learning things... π
Hi Guys, new to HA and I'm struggling a little. I've got HA running in a VM and can access it via my browser. I've got my google minis showing and working so that's great. I've installed a few addons (terminal, MQQT broker). I'm now trying to add something from github that I cant see in the HA panel (https://github.com/matijse/eufy-ha-mqtt-bridge). Can anyone link me to a resource that tells me how to do this (i've tried something called HACS but I get an error trying to add a custom repo).
@viscid orchid
β Thanks
@late sierra You are trying to install the add-on?
If so, you would need to add the add-on repository to your add-on store
I don't see any obvious way of adding the GitHub repo in the ui
upper right corner of the add-on store page
And for future reference, HACs is for custom integrations and UI cards. Nothing to do with add-ons, which are software services.
So here, you add MaxWinterstein's add-on repo...
And you get that repo added to the store:
And then you would proceed to install and configure the eufy add-on
@late sierra did you get it?
Ace thanks mate, I'm out at the moment but will try it and report back when I'm home
Odd following the https://github.com/MaxWinterstein/homeassistant-addons install instructions (same as you explained) i'm getting an 'Invalid Add-on repository!' error
Works for me
As mentioned, has nothing to do with HACS, so if youβre messing around in there youβre in the wrong place
No I was in the correct place doing the correct things. Just got it added. My solution was to restart HA then follow the procedure, no idea why that solved it
hello everyone. Does anyone know if there is an add-on to have home assistant act as a tftp server.
After updating Ha. node red cannot connect any more to HA. This is my error; 1-05-14 19:26:23 ERROR (MainThread) [supervisor.api.ingress] Ingress error: 400, message='Invalid response status', url=URL('http://172.30.33.3:8099/socket.io/?EIO=4&transport=websocket&sid=DBaa0YcaI0_6t3bsAAAA')
Hey guys. For the portitainer add on in HA OS, how do i get the option to add a stack?
Hi all, new user to Home Assistant here, is this the right channel to asking about the DuckDNS addon?
yep
It's more of a question around the process of installing a cert using the Duck DNS addon - It might not be the right place, but you folk may be able to answer my question. I'm using the Everything Smart Home Tutorial for installing a certificate with the DuckDNS addon. The video talks about setting up port forwarding specifically 8123 -> 8123 and 443 -> 8123. I have just discovered a limitation with my router where I am unable to port forward two externals to a single internal port. Will this be a problem with installing a certificate with DuckDNS?
I have searched around home assistant community and google and I haven't really found anything concrete
@delicate anchor you don't need both. choose one or the other
8123 to 8123 means you'll have to include :8123 at the end of your duckdns url
443 to 8123 means you don't
right, right, understood - awesome thanks for your help @sly kindle I will go forth and conquer!
Hi guys. I spent few days finding the answer, and here is my last hope. I installed an add-on "Home Panel" on HA installed in VM, but I can't figure out how to use it on my wall tablet.
I've created a beautiful dashboard (great job, @timmo!), but I can access it only from HA, and I can't get rid of the top line. I always have to log in to it, and each time I have to log in to HA.
I read somewhere that I can access directly to HP via https://x.x.x.x:8234, but I never made it.
Can you please guide me on where to find a step-by-step setup?
Thanks
Sometime I get "Invalid Add-on repository" trying to add it to the add-on store, why ?
Youβre probably trying to add something that isnβt an addon repo, like something intended for HACS
How do you save in the VSCode addon?
it auto saves. ctrl + s or cmd + s should also save beyond that
But on mobile?
it auto saves
Can that be turned off? So easy to mess up on mobile with the keyboard popping up all the time
I mean, turn off autosave and have an option to save with something else π
stop editing on mobile π
I was hoping it would be better than File editor which is impossible on mobile
itβs also easy to correct as long as you do not restart with the messed up code? π§
use the nfs:// or smb:// or afp:// or whatever addons are out there to access the files from your mobile, use a app to edit
Yeah, but that doesn't work properly always either. The editors work fine but saving can be problematic sometimes.
Anyone have any idea why tasmoadmin keeps failing with
Logger: homeassistant.components.hassio
Source: components/hassio/websocket_api.py:109
Integration: Home Assistant Supervisor (documentation, issues)
First occurred: 2:51:37 (3 occurrences)
Last logged: 2:56:46
Failed to to call /addons/a0d7b954_sonweb/stats -
I need to run a vscode extension that needs administrator perms but the vscode add on doesn't run as administrator
My possible solution would be to run the docker container for it as another user
A quick question (and apologies if asked before, I've searched around, including searching through this channel itself, but haven't found much useful info). I'm a few weeks into running the HASS OS (previously I was running HA via venv) and while I'm liking it, I can't figure out how (or rather, where) persistent volumes are stored when using the official portainer addon on the host system. I can't use something like ~/docker/[container_name ...] for a bind mount as it's R/O. Though volumes do work and data is persistent through container redeploys, but I cannot find the actual location of them on the host system, if that makes sense. They're not stored in /var/lib/docker like other distros and the mount point listed under portainer leads to a directory that doesn't exist on the host system. Anyone ran into something similar?
@cold cradle - for my setup, the mount point on portainer: "/mnt/data/supervisor/" seems to map to the root (via the ssh add-on). That's at least where I am going to manage my addon config files on the host (/mnt/data/supervisor/config = /config)
Indeed, I've located the config for Home Assistant itself. I figured there was some odd mounting going on in the HASS OS though, likely for security purposes. I did see that folders like /addons, /share and so on are symlinks to folders buried within numerous different subfolders and I did have a look around them a few hours ago but didn't find anything remotely related to portainer or its data. Unhiding the supervisor/core containers and having a look at their configs did reveal, as you mentioned, about the /mnt directory but that's empty as it seems. There are a bunch of 'devices' within /dev and I presume they hold the data for the HASS install itself, such as core, supervisor, dns etc. I guess it's a side-effect of using portainer as an addon to HASS -- that they're also stored within those sda 'devices'? I honestly have no idea. It's not a huge issue right now, but I prefer bind mounting important container data so I more easily perform backups on it, amongst other things.
Does anyone have experience with the wyze sense component? I am having a pairing issue pls @ me if replying
not an addon
anyone know how to fix this
maybe you can describe what extension specifically? afaik most containers run as root
It's a custom made component I made for vscode. I don't want to publish it yet but it allows for easy editing of automatic code replacement programs. It needs administrator perms because it edits core files of vscode.
this might be a question then for generic vscode containers, not HA.
but as said, majority already runs processes as rootβ¦
when started the vscode addon warns code-server is being accessed over an insecure domain. Web views, the clipboard, and other functionality will not work as expected. Is this the case? Do I need SSL for it to function fully?
yes, you need ssl. itβs more of a browser limitation than vscode.
Hello my deConz keeps crashing after last update to 2021.5.4
any solutions ?
running as an vm ware template
I had no backup selected any way to downgrade ?
hmm.. I just restored a snapshot of the lets encrypt addon on a new home assistant system and it restored more than just the addon and its configuration. it also restored all the custom addon repositories.. is that supposed to happen?
hi,
i want to implement the https://github.com/pbkhrv/rtl_433-hass-addons repo.
but if i want to insert it in supervisor, nothing happens.
what am i doing wrong?
@real seal adding it to the add-on store page works here
@raven sparrow ha core update --version=2021.5.3
i have Core Version core-2021.5.4
adding other repos works
it just happens nothing if i klick "add"
solved
i restarted the whole server and now it works
strange
other question: is there an add-on fΓΌr the nano-cul?
nano-cul?
oh, its a usb-serial-433 device
you just try to get it running with proper drivers and pass through the proper resulting device to a generic 433 container
@sly kindle i have the same issue as @raven sparrow , downgrading to 2021.5.3 didn't help
@hoary tideI was only sharing how to downgrade, really. I don't know that this would resolve anything.
oh okay. i'm trying going back to 2021.4
@hoary tide it sounds like it may not be related to the HA version. You might check the deconz discord server
i'll try thanks
i made it work again by updating homeassistent to latest, uninstalling deconz 6.8.0 and restoring deconz 6.7.2
Is there a way to access an addon that uses ingress by a separate URL to break it out of the HA "frame"
in my case, grafana
If you lose password for the nginx addon how you reset it?
Asking for a friend cough
oh i think i may see the problem.. Nope :/
you can set a port in the addon config tab
in the case of grafana you need to also set to allow anonymous auth, then create your own user/pw in grafana
obviously it will no longer work with ingress auth from this point on
When a snapshot is made of an addon is it's database also snapshotted in that? Or do i need to restore the entire mariadb for that?
if you use the mariadb addon, yes the db is included in the snapshot
ok fiddling with db a bit i got back in π pfew
Good morning I installed the grocy addon this morning and according the logs everything is running great. Unfortunately the install appeared not to create the grocy sensor as described in the Read Me. Did I miss something? Anyone else having this issue?
for some reason i don't see any of the things i set up in nginx.
https://i.imgur.com/Q2Vxwrw.png
But it appears to be functioning. Any clues?
I have permission for all-items, am admin, and none of the manage items are Hidden. Under permissions for Administrator. At some update was there a move from file storage to DB maybe> All my tables are empty in the DB for the proxy_host, Redirection_host tables. I used to be able to see them.
hmm restored march backup, same problem.
wonde rif it happened when i migrated from Pi to NUC and i just never noticed it before.
Guess i just tear it all down and set it up again :/
your npm db seems empty, your nginx rules stayed though if it works
yes, i have already uninstalled though and am setting up again.
uh yes 1 added, and retained in NPM
so am i understanding correctly the two are not linked?
NPM just pushes to nginx but doest realy know what's in nginx config?
yes, it's only one way
Very convenient tool though. took me a few mins to set up 3 proxies and a 404. With SSL.
But i relay just went looking into it coz ISP gave me a new modem so i am expecting my IP to change and i hoped there was some stuff in it for changing DNS automagically π
There isn't. But maybe my IP will remain fairly static and it won't be an issue. I am being moved to dynamic IP π¦ π¦ π¦
some modems actually have that integrated too
Somehow i doubt my ISP's router will have that. π
But ill find out soon enough.. gonna install it now
GM Everyone due to a stupid mistake on my part, my HACS install got all bjorked. I've attempted to remove it using the uninstall instructions and then install it back in, but when I do that nothing for it shows up. Is there a way to completely purge this thing from HA so I can get it clean?
hacs is #integrations-archived
nvm
Taking a look at the hassio-addon structures.
How does hassio know where the config file is?
Is it recursively looking through all directories for the necessary file?
or is it addon-[example] so config files must be located in [example]
Thank you π
Hi all. I have the nginx and the letsencrypt add-ons. Every once in a while, the certificate gets renewed correctly, but nginx doesn't get restarted, and after a few days I end up with an expired certificate error. Has anyone here solved that issue?
Is there some kind of event I can hook into when letsencrypt renews the certificate?
Hmm, I think I've managed to automate it. A bit creaky.
How long does it usually take to install an add-on?
Because it looks like my homeassistant can't install ESPHome
After ten minutes it finally installed.
i need help with the following please:
I installed HASS (with MQTT, NodeRED, etc.) on a phone using this tutorial (https://github.com/alexanderhale/home-assistant). Now, after finally converting my ceiling light from tuya to tasmota i don't know how to set up mqtt. it's, just like my flashing, the first time. I haven't set up a password or a username. i am running hass and mosquitto separately (i think?) on termux if that helps.
Hi all. Is this an appropriate place to ask a question about Node Red AddOn?
Thx.
So I just updated the Mosquitto add-on to v6.0.0 and it no longer loads my bridge configuration located in /share/mosquitto. This breaking change was not mentioned in the release notes, so how do I go about fixing it?
@raven tinsel
Please, raise an issue on the issue tracker with a reproduction scenario
done. #2011
I have rolled back for now.
Hi everybody.
I am trying to add a second UPS to be monitored using hassio and NUT. right now I have one UPS beeing monitored but I have a second one that it shows me the same info as the first one.
I tried to add a second one in the configuration just only changing the name, but it doesnt work
Any idea?
Thanks in advance
I have hass OS, with the ESPHome addon, but in order to troubleshoot an issue i want to be able to upload firmware without recompiling it - I understand i can do this from the CLI of ESPHome, but how do i access that if i have home assistant OS?
Is there an add on or something to get RGB lights to sync with audio coming from my computer?
use emulatedhue and the hue sync app on your computer would be the simplest i guess
hmm, would emulatedhue also work with Razer Chroma?
if it's registered as 'light' in HA, yes
Did someone ever made a servo work with esphome?
@willow lintel there's a new version of mqtt add-on to try
apparently not, 2 of the lights i have (that are actually registered as lights) are incompatible with Hue's "entertainment area" mode
emulated hue should make all lights compatible
ok looking at it, apparently HA has those lights registered as not having color control
maybe something else that can help me, is there any sort of add on to flash RGB controllers?
sounds like it's going to be very specific to the controller?
i guess something like tuyaconvert
oh so you're talking about flashing them with new firmware
yeah
it'll depend what kind of chip they're running on, not my area of expertise
I guess if need be, i can grab a spare SD card and set it up for tuyaconvert
any idea how to add a second UPS de NUT addon?
I dont know how to do that. I am running Hassio in a pi4
one proper way would be to run #449717345808547842 and learn a bit of docker
the hard way would be to try run it on hassos and tools like portainer
why do you think that it doesnt work just adding a second one in the configuration file?
I have no idea about the addon, don't use it
if you don't even provide your config - how do you suspect getting help works?
well the documenation has examples of multiple devices
Hi guys, since an about an hour my Mosquitto broker gives me an Error "Error: Address in use.". What could cause this? I already restarted all my networking equipment and the VM HA is running in as well as the Addon serveral times by now. Nothing helped so far, hopefully you can help me out.
Just found this is my Supervisor Logs: 21-05-19 16:41:49 ERROR (MainThread) [supervisor.services.modules.mqtt] There is already a MQTT service in use from core_mosquitto
Is there a second MQTT Service? I don't quite get the message
How do I mount a drive in an add-on? I'm running HA supervised and would like to mount a second hard drive to store motionEye recordings.
@keen tendon You don't
Add-ons don't support mounting additional drives
motionEye does offer capabilities by itself though
I can store it on a network share. But the only drive I see is my nvme drive.
? That response doesn't make sense to me
I wanted to share a screenshot, but doesn't work.
it can store on a network share
There are three options, custom path, network share and KINGSTON NVME drive.
what you expose through a network share (nvme or smoke signals) doesn't matter
I could of course start a samba share on the host and point motionEye there.
But I'd need to add a USB -> smoke signal dongle...
hehe π
but external drives / additional drives
that won't work
network shares it can do
I'll give that a try. Thanks Frenck! π
Anybody know how to sideload esphome firmware? I want to compile it here, then transfer the firmware to a remote home assistant instance and upload to the node from there
because the remote home assistant crashes every time i try to compile it there
Can't you simply put it in the same folder on the other machine and then press upload?
locating said folder is the tricky bit
I think it should be in config\esphome
I have a bed scale. I can find the .yaml file in config\esphome, and I think all the buildfiles are in \config\esphome\bed_scale
the firmware file isn't, one of the config yaml files states the firmware is in data/.pioenvs/
Ah yeah. Found it on my file.
For me it's in /usr/share/hassio/addons/data/XXXXX_esphome/bed_scale/.pioenvs.
Basically it's the folder higher than your config directory. Then addons/data/...
On the other hand, maybe you want to look into why HA crashes while compiling?
doesn't look like we have the same structure here
I run supervised.
I'd love to but it's difficult when people prefer "catch-all"
excuses to actually investigating an issue
it's not the PSU and it's not running out of memory, but those are the only options anyone is willing to explore
outside of that and "MUST B UR FAULT NUB"
What's your install method?
it would be /mnt/data/supervisor/addons/data/...
but you have to be connected to the host console
and not using an add-on
is this something i can do without being physically present?
Depends on how long the cable of your keyboard is π
no. you need to be present to either connect keyboard and monitor, or to import your ssh key to use debug ssh
(import from a USB key)
can you upload it directly to the node?
so currently when i ssh to the box, i'm hitting the addon ssh
yes
so do i need to prod a different port to get the host terminal?
as i understood it, without the addon, you just get no reply
You'd need to SSH into the host I think.
Sorry to interrupt your convo, but I found something again, maybe HA is having issues checking my ports? https://pastebin.com/NrUqQKzx (from Supervisor Logs)
thanks ^ would it be naive to think that opening the relevant port on the router at the other end and using key based authentication is pretty solid?
as solid as ssh and keys is π
an attacker would need to be literally in my house or intercept the key?
the problem is you need to be present to import the key
i can do that, i visit on occasion
and i'm guessing that's intentional because of the whole.. intercept the key thing
the good compiler is here, so that's all good π
it's getting it onto the remote host that's the issue, but it looks like i won't be able to do it until i can visit the site again
@thick herald did you update the mqtt broker ?
oh, unless i can upload it somewhere else and change the config that tells esphome where the firmware is? oO
Maybe it updated automatically, that option was enabled, currently reapplying a snapshot from earlier this day
you would definitely want this latest version
it's working again after applyping the snapshot, I'm checking the verison
@waxen plume is ota enabled on the node?
It's saying it has an update available which it didn't before
Maybe that broke it for me?
@waxen plume
I'll update (again?) and see what happens
Dining table Web server is something I didn't know I needed π
if you give it a simple config without whatever is causing you your compile issue...
it's a light above the dining table π
I can confirm the update breaks it
what is your add-on config @thick herald
not sure that cafile: is an option
Problematic Log Output:
https://pastebin.com/fvZNNwXG
It says in the documentation
I might use it wrong though π
that ca file is wrong/cannot be right
see π
It was working before, I'll change it real quick
ok
Removed it, didn't fix the issue unfortunately
what issue?
Since the update to the latest version of the addon I get the above log output
that has been fixed, but requires a new Supervisor to be shipped
you can stop it from appearing by disabling "watchdog" on the add-on
Will it fix the addon as well? it's not starting the broker as of now and my zigbee2mqtt can't connect to the broker
or will it just disable the log output
Will it fix the addon as well?
That error posted is not related to the add-on
and cannot be "fixed" by the add-on
so, you have another issue than?
Looks like it, if I get your description, can you help me identifying it? It worked before updating the addon to the newest version
my description?
could you just share the error log from the add-on?
you say it doesn't start
It's in the pastebin above, I can add all the log output, that would probably help, yes.
The pastebin above is the supervisor log
not add-on log
as said, that in that log is not an add-on issue
Oh i see
1621445245: Opening ipv4 listen socket on port 8883.
1621445245: Error: Address in use
Here's the complete log: https://pastebin.com/7awfQkw1
Seems like you are loading custom configuration
anything in there that tries to address the same ports again?
1621445833: Loading config file /share/mosquitto/mosquitto.conf
well I didn't install it or create the custom config... So, I cannot know either π
the .conf file only specifies the 8883 port
is the file conficting with the addon config in the UI?
Here's the file contents: https://pastebin.com/jG6di2Y2
You don't need that :S
ok, just delete the file or the config option for it?
that is only causing issues
you can delete that file and disable the customize active to false in the add-on config
Ok, we're further, I now get this message here: 1621446142: OpenSSL Error[0]: error:1417C0C7:SSL routines:tls_process_client_certificate:peer did not return a certificate 1621446142: Socket error on client <unknown>, disconnecting. 1621446143: New connection from 192.168.2.1 on port 8883. 1621446143: OpenSSL Error[0]: error:1417C0C7:SSL routines:tls_process_client_certificate:peer did not return a certificate
That means you have require_certificate enabled
but the client connecting there, doesn't send one
Wow, all the months I had a wrong config, just didn't notice it until today
thanks, it looks like everything is working now
π
@covert arch
here is my NUT conf with one UPS and working
https://pastebin.com/GhFpjgwG
and here is when I add a new UPS to the conf and have the error
any help?
maybe is simple to fix it
both UPS are the same
Thanks in advance , I am jr with this π¦
only warning left is " running with disabled protected mode!" for portainer but that can't be helped i think
Most of the tim eits not running anyway
greetings! A few versions ago, around February 2021.. I lost access to my zwave front door lock from the Home app on my iphone/ipads. It no longer appears on any of them. The zwave lock works fine from within home assistant. I am currently on the latest version of core, 2021.5.5. Any suggestions on how to restore access to the lock from iOS home app?
I am using the homekit add-on, with an appleTV 4K as the home hub.
All the other devices are showing up correctly in the Home app, and I do have less than 150, or whatever the number is for the limit
I am using hassOS on a raspberry pi 4b with 4GB ram, and using 200MB of that
the lock is a kwikset convert 914 z-wave plus
Hi! After the last Mosquitto update, PAI (https://github.com/ParadoxAlarmInterface/pai) has stopped updating its MQTT topics. It seems to connect fine, I get a successful connection message in both the PAI logs and in the Mosquitto log, and the debug logs of PAI show it sending messages to MQTT, however the topics in MQTT never get updated.
Any tips on how I can troubleshoot this? I'd like to try downgrading the Mosquitto addon, but I'm not sure how.
there was a discussion about Mosquitto earlier today: #add-ons-archived message
Ah yep I saw that, I have disabled the watchdog and restarted both Mosquitto and PAI, but no change
@sullen musk - I think it's HA not talking to MQTT - not PAI
I'm having the same issues
I checked using MQTT Explorer, and PAI is definitely not creating/updating topics
My supervisor logs are saying:
21-05-19 23:41:20 INFO (MainThread) [supervisor.auth] Auth request from 'core_mosquitto' for 'homeassistant'
21-05-19 23:41:20 WARNING (MainThread) [supervisor.auth] Unauthorized login for 'homeassistant'
2021-05-20 09:43:20 ERROR (Thread-3) [homeassistant.components.mqtt] Unable to connect to the MQTT broker: Connection Refused: not authorised.
2021-05-20 09:43:20 WARNING (Thread-3) [homeassistant.components.mqtt] Disconnected from MQTT server core-mosquitto:1883 (5)
but there doesn't seem to be anywhere to CHANGE/RESET the Homeassistant user password
as far as I can see - PAI and everything else talks to MQTT properly, but HA now isn't
HA seems to be writing to MQTT fine, if I trigger an action in HA it create/updates the correct topic. So I think I'm okay from that front.
hmm - ok - we may have two separate issues with the same integration...
mine at the moment is - everything else will log into MQTT, EXCEPT HomeAssistant
@night bluff Remove MQTT integration restart HA
MQTT should be discovered
add it
β
I'm using HA Supervised
so?
ok - I'll try it again - one sec
Mosquitto Broker in Supervisor has been uninstalled
I didn't say that
please read my instructions again
as you have now removed the add-on
I said: Remove MQTT integration
ahhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh
fucknz;lxkjbgvl;ksdjbgl;ksjdbgvjsdfbvjsd.;
ok
I'm apparently an idiot this morning
10am here - I have zero excuses
ok - RE-ADDED and re-added the config for MQTT in Supervisor
THEN REMOVED integration (Duh!) from HA and am restarting HA
hmm
yes that is expected
ok
as you removed the integration, so add the integration back π
it's discovered it
hitting configure
wants another restart
hmmm
ok
so now the MQTT log is only showing ONE connection - and thats Homebridge.... I'm not seeing PAI or HomeAssistant or Node-Red connecting....
That is odd, should be normal. Home Assistant discovered it...
and I'm still getting :
1621470554: New connection from 127.0.0.1 on port 1883.
1621470554: Socket error on client <unknown>, disconnecting.
that means something is connecting with incorrect info
ok, emulatedhue is constantly erroring with
2021-05-19 20:02:47,843 ERROR aiohttp.server -- Error handling request
Traceback (most recent call last):
File "/usr/local/lib/python3.8/site-packages/aiohttp/web_protocol.py", line 422, in _handle_request
resp = await self._request_handler(request)
File "/usr/local/lib/python3.8/site-packages/aiohttp/web_app.py", line 499, in _handle
resp = await handler(request)
File "/app/emulated_hue/api.py", line 90, in wrapped_func
return await func(cls, request)
File "/app/emulated_hue/api.py", line 492, in get_full_state
"scenes": await self.async_scene_to_full_state(),
File "/app/emulated_hue/api.py", line 478, in async_scene_to_full_state
del scene_data["lightstates"]
KeyError: 'lightstates'
Is there a way to rollback an addon to a previous version?
Restore it from a snapshot
mm ok I will see if I have a snapshot with it., thanks π
it worked :D, the only issue now is that is't a 10 weeks old version haha
hmmm
ok - the new version of MQTT seems to be doing weird shit
PAI isn't talking to it anymore since it upgraded this morning...
clarification - PAI isn't talking to MQTT since MQTT's update this morning....
6.0.0 -> Support for anonymous logins has been removed
is it trying to use an anonymous connection?
nope
checked that already
trying a rollback (hoping the snapshot I did this morning was BEFORE the MQTT update!!)
PAI?
Paradox Alarm Interface
and sadly - my snapshot was only for the Core update - I'd done the MQTT update PRIOR to that and stupidly said "nah - I don't need no stinkin' backups!"
facepalm
Pai says:
2021-05-20 16:17:55,186 - ERROR - PAI.paradox.interfaces.mqtt.core - MQTT Broker unexpectedly disconnected. Code: 1
2021-05-20 16:18:03,396 - WARNING - PAI.paradox.interfaces.mqtt.core - Connection failed, retrying: Connection refused(111)
In my case PAI was reporting a successful connection, but just no communication was happening. I took the advice of rolling back Mosquitto from a snapshot, which has solved the problem. After rolling that back I've noticed that zigbee2mqtt was also not communicating properly.