#add-ons-archived

1 messages · Page 9 of 1

obsidian pawn
#

Illegal instruction (core dumped) ... I think the add onn was corrupted. I uninstalled and re-installed. It came back, now I am trying to figure out if my DB is still there

weary stratus
#

Hey there,
on Hassio, I get this error, whem checking for DHCP Servers in AdGuardHome.

Hassio got a static ip.
Any ideas what to do

#

I am using the servers network interface. not hassios or adguards

obsidian bluff
#

Quick question: do home assistant add-ons have their own backup functionality for their own data, or is that somehow provided by HA itself?

bitter skiff
weary stratus
#

NEED HELP. I just moved from (dieing) SD to SSD by making complete fresh install on SSD an restoring FullBackup from SD. Worked like a charm except my Add-Ons. None of them got restored. Add-On Menu is empty. Installing and configuring them all new will take hours...

#

Made some (hard) reboots..doesn't work

still kelp
#

Hi all, have home assistant running on an old raspberry 3 and try to install MQTT. Installed the Mosquitto Add-On, added a new local user for this purpose, saw the MQTT integration and configured it. Is there any way to test whether it is actually working?

#

"Another way to send MQTT messages manually is to use the “MQTT” integration in the frontend." which seemed to work. I have an ai-on-the-edge ESP running but nothing happens when I try "MQTT discovery"...

weary stratus
pallid elk
#

Hi I just wondered if anyone has seen an issue when trying to install an addon like this:

Can't install ghcr.io/mcinnes01/amd64-addon-grafana-loki:1.0.0: 500 Server Error for http+docker://localhost/v1.41/images/create?tag=1.0.0&fromImage=ghcr.io%2Fmcinnes01%2Famd64-addon-grafana-loki&platform=linux%2Famd64: Internal Server Error ("Head "https://ghcr.io/v2/mcinnes01/amd64-addon-grafana-loki/manifests/1.0.0": denied")

I'm running a supported supervised install on debian and have plenty of addons installing fine, but I'm noticing this on a few. I acutally forked the original version of that addon and built the artifacts in my own github registry to verify if it was just that the registry was private but weirdly I can download the docker images no problem without logging in so it seems like they are public as expected.

still kelp
#

Seems that I solved my MQTT issue... No idea how...

dusty granite
#

Hello, I've been looking around quite a bit online but can't seem to find the answer. Please don't tell me to install supervisor or OS, it doesn't suite my use case. I want to know how I can install/use add-ons for Docker version of HA. I'm specifically trying to get the cloudflared add-on to work

supple grove
#

You can't as an addon is a docker container build specifically to work with HA. You can however install and configure the software it uses yourself. They're mostly available as container as well

dusty granite
#

Hmm interesting, seems like it's been done in the past. Looks like I can run the add ons as seperate containers and configure them to work together

stiff grove
#

Getting an error trying to run AdGuard home telling me port 53 is already in use. How do I see what is using port 53 and fix this?

supple grove
#

netstat. You can also try nmap from another host not it might nit be bound to the outside. I'd start with docker ps it's probably a container using it and HAOS isn't fun to troubleshoot

stiff grove
#

It was adguard running itself. Uninstalled and reinstalled and it started up by its not functioning correctly

livid atlas
#

I have a problem with tailscale addon.. i wanted to use in combination with remote-home-assistant to link 2 instances located in different houses, however on my main node i'm not able to start tailscale addons.
Tailscale dashboard says node is online but addon on homeassistant doesn't start.
Thanks to anyone in advance that could help me

supple grove
#

Check the logs

livid atlas
#

Only error i can recognize is related to "dns" and "dns-os"

supple grove
#

Can you share them?

livid atlas
#

2023/03/06 11:57:59 health("dns-os"): error: getting OS base config is not supported
2023/03/06 11:57:59 health("dns"): error: getting OS base config is not supported
2023/03/06 11:57:59 control: controlclient: restarting map request for "dns" health change to new state: getting OS base config is not supported
2023/03/06 11:57:59 control: controlclient: restarting map request for "dns-os" health change to new state: getting OS base config is not supported

shy stone
final jay
#

I have a question about a custom lovelace card called "Config Template Card" card. I'm attempting to add it as a resource but the documentation says to install a javascript file but the card's repo only contains a typescript file. Any guidance on how to proceed would be much appreciated! Repo in question: https://github.com/iantrich/config-template-card

viscid orchid
#

You need to look at the release

#

And you'd be better off installing via HACS

river wharfBOT
final jay
#

Gosh, it's so easy. Thanks.

#

I looked for it on HACS but couldn't find it. I guess I'm a bit inexperienced with it as well.

#

Ah wait. Perhaps I have to add it as a repo...

#

Found it. Thanks for being a rubber ducky.

dusky sparrow
#

did you find a solution? My Yellow is on the way, and I wanted to do the same...

viral fractal
dusky sparrow
viral fractal
# dusky sparrow Perhaps HmIP-RFUSB works? no idea...

My research so far indicates that a HM-MOD-RPI-PCB connected to J11 should work „out of the box“. Same with the USB variant. But it does not work for me as it is not getting detected and I have no possibility to configure anything. Maybe I am missing something. If you learn something new please let me know 😊

surreal basin
#

I have a question about setting wireguard and nginx, ssl, etc... Been stuck for hours googling and maybe I'm overthinking

#

I cannot access through wireguard VPN to SSL connections such as my vaultwarden or homeassistant app. Nginx seems to not work for me and i don't know how to fix it or if I should just not use https connections

bitter skiff
#

Any idea if it's possible to get the logs of a restarted add-on? One of my add-ons was restarted by the watchdog and a standard docker logs <container_name> only yields logs from the currently running container.

livid atlas
misty yarrow
#

Hey, this is quite niche so pretty low expectations for support here but I'm struggling as someone who is very new to python. I've forked this add-on to try and push/pull messages over UDP and MQTT to a smart immersion heater. The author of this repo set it up so the IP and MQTT topic for the device is hard-coded. I have two devices, so need to be able to parameterise the devices. I've had a crack at this but it's not working. If anyone feels like having a look through the repo (and my fork and letting me know what I might be missing that would be much appreciated!!!) https://github.com/benholl/tsmart/tree/multiple_devices

misty yarrow
wild basin
#

Hello

#

I tried both SSH modules on HA it all gives me the error: 502: Bad Gateway

obsidian stratus
#

Hi there. After updating HA to 2023.3.1 VScode addon doesn't load anymore when behind nginx reverse proxy. I get this in the logs: https://paste.debian.net/plain/1273291. Has anyone seen this before?

strong tendon
#

It’s a long shot but has anyone managed to block all4 ads using either adguard or pihole none of the blacklists online work for all 4 please @burnt glade with reply 🙂 thank you

sharp marten
#

Has anyone had this issue with the mosquitto mqtt add-on? This was previously running just fine and I don't think I had changed anything before it stopped working. I have since updated everything, and it is still broken:

2023-03-03 21:24:04: Opening websockets listen socket on port 1884.
2023-03-03 21:24:04: Error: Unable to create websockets listener on port 1884.
[04:24:04] WARNING: Halt add-on
azure crypt
#

Umm, what's the difference between the "terminal & ssh" addon, and the "ssh & web terminal" community addon?

supple grove
#

Different functionality and default settings

dawn scaffold
#

Pi-hole
service call pi_hole.enable does not exist. pi_hole.disable, however, works just fine. What up?

lofty solar
#

NVM, it was a Firefox issue
edit: that's what's happening when "privacy.resistFingerprinting" is enabled

plush coral
#

Hi, I have configured Alexa Skill on HAOS, but filters doent work, somebody knows why? I think yaml files are correct

sly kindle
surreal basin
#

Can someone help me to use my wireguard with hostnames instead of ips? I also have adguard but can't get to make it work. I need to type ips to connect with wireguard and I'd like to use hostnames. My wireguard server has adguard dns configurated but I believe i need to use reverse dns or something?

void prawn
#

I'm having trouble with my AppDaemon app I created to notify me about doors being open. I want to create something streamlined and simple that notifies when any door is open and then also notify after a certain amount of time, but I want to be able to control the amount of time between notifications and I want to be able to set different times per door. I have input my python, YAML, and the log into a pastebin here:

https://pastebin.com/2xrbZVrd

The devices are reporting correctly, they change in my HA UI and they report to the log when they open and close. I am just not getting any notifications pushed to my phone like I used to with the old file I was using and I don't know why this isn't working.

bitter skiff
#

You're never going to get to line 37.... tracker(...) is only called upon state change, so when it's on it'll create a "timer" and then proceed to never do anything with that value since there isn't an actual timer set like you've got in line 21 of your old version.

void prawn
#

Ok. . . . so it is starting the timer, but not putting that anywhere so it basically just stops there?

#

Can I add line 21 form the old version into the code of the new version?

bitter skiff
#

You should probably update your new version to have the timers map created similarly to your prior door_timer_library. In the new version, it's just a map to a datetime, but the old version actually stores a handle to an AppDaemon timer

#

That timer is what actually executes the notify if it's elapsed the ttl

void prawn
bitter skiff
#

At a minimum you'd want to indent line 34, otherwise it's going to start 2 timers

void prawn
#

I have updated accordingly. Is this all I should need for now to get this working?

P.S. You are the best, by the way!!!

bitter skiff
#

I can't thoroughly review it at the moment, but that update will at least get notifier triggering after your TTL

#

So progress was made

void prawn
#

Ok.. I will test and see! Thank you for your help!!

#

So far no error. Now to test in about 30 minutes when I head out the door

wispy kestrel
#

Hi - I'm new here. Have been using HA for a few months. I'm trying to get multiscrape to pull lunch options off my kids' public-facing site. I think I have the right css selector and have downloaded multiscrape in HACS, added the sensor to my configuration.yaml and restared HA (multiple times), but I'm not seeing the sensor in my entities. Is this the right place to ask for help?

bitter skiff
#

Sounds like an #integrations-archived, not an addon. You can post there w/ your config and make sure you check your home-assistant.log as well to see if there were any errors loading that platform

wispy kestrel
#

@bitter skiff - thanks. Is the best way to just search /config/home-assistant.log for multiscrape? (haven't spend time in logs since I've started w/ HA)

bitter skiff
#

Yeah, I'd start there. Depending on how noisy your log is... you could just try restarting HA and taking a look, the log rotates when you restart so it should be pretty short if you don't have loads of debug logging on

wispy kestrel
#

I see it:

2023-03-08 14:08:23.760 ERROR (MainThread) [homeassistant.components.sensor] Error while setting up multiscrape platform for sensor
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/helpers/entity_platform.py", line 293, in _async_setup_platform
await asyncio.shield(task)
File "/config/custom_components/multiscrape/sensor.py", line 57, in async_setup_platform
if not coordinator.last_update_success:

Not sure what to do with this, unfortunately. But, appreciate I'm in the wrong spot, so happy to take this to #integrations-archived

bitter skiff
#

If you could paste that full message (im sure theres more to it) to something like pastebin and share the link over there, that'd be great

wispy kestrel
#

Thanks!

slender karma
#

hi, concerning the MariaDB addon, is there a way to connect to the mariadb server from outside HAss ? I'm trying to connect from my PC to this db

bitter skiff
#

sure

#

What have you tried?

#

Make sure 3306 is filled in in the config tab

slender karma
#

I've set as host my hass ip and 3306 as default port but i get a connection refused

bitter skiff
#

In your Mariadb addon config page

#

Ensure 3306 is defined as an exposed port

slender karma
#

sorry i haven't seen the textbox on the left to set the exposed port

#

working like a charm !

void prawn
bitter skiff
#

Have updated logs?

river wharfBOT
#

@void prawn I converted your message into a file since it's above 15 lines :+1:

📎 message.txt
bitter skiff
#

If 15:21 after we made the changes

void prawn
#

yes

#

my doornitfy log is showing the sensors changing state correctly

2023-03-08 15:15:10.341406 INFO door_notify: binary_sensor.upstrs_back_door_sensor has been opened.
2023-03-08 15:15:17.158372 INFO door_notify: binary_sensor.upstrs_back_door_sensor has been closed.
2023-03-08 15:21:52.678528 INFO door_notify: binary_sensor.garage_entry_sensor has been opened.
2023-03-08 15:21:55.860276 INFO door_notify: binary_sensor.garage_entry_sensor has been closed.
2023-03-08 15:22:11.488851 INFO door_notify: cover.garage_door has been closed.
2023-03-08 15:22:56.503470 INFO door_notify: cover.garage_door has been closed.
2023-03-08 15:23:11.478540 INFO door_notify: cover.garage_door has been closed.

#

It is strange that the cover.garage_door is notifying 3 times in the log when it closes. It isn't logging the open at all

bitter skiff
#

Does it perhaps not have an on? Is it maybe open?

#

(I don't have any covers, not sure)

void prawn
#

That's a good question. It may be open instead of on. . . and that's fine for now. If I can get the notifications working on this python file I can fine tune that issue after that.

bitter skiff
#

for your time_since_changed.... you can just have it be return self.get_entity(entity).last_changed_seconds

#

that'll fix your latest error

#

Well that'll return seconds, / 60 if you're looking for minutes.... line 46 shows you're converting the TTL arg into minutes

void prawn
#

I want seconds when less than 60 seconds and minutes when greater than 60 seconds

bitter skiff
#

That sounds confusing 🙂

void prawn
#

I just don't want to see a fraction of a minute, when it is in seconds

#

if you have a better way to make that happen, I am all ears. I am not married to any of the code, so I don't mind changing anything if it is better.

bitter skiff
#

Well... how would you see you knowing whether 5 means 5 minutes or 5 seconds then

void prawn
#

I also don't want to just have it in seconds and see 120 seconds instead of 2 minutes.

bitter skiff
#

Having a single value mean two different things isn't a great idea

#

You could have ttl_min and ttl_sec if that's really desirable, and just ensure you set one of them and use that

void prawn
#

I see what you are saying. I did have it set up differently and I rely heavily on AI as I don't know python very will. I can read it minimally, but that is about it.

bitter skiff
#

I'd recommend continuing with what you have currently before adding more features 🙂

#

Right now, it's in minutes

void prawn
#

Ok. I changed it to read: return self.get_entity(entity).last_changed_minutes

bitter skiff
#

can't do that

#

the property is named last_changed_seconds

#

You need to do that and just divide by 60

void prawn
#

Undone

#

how do I add the divide by 60?

bitter skiff
#

return self.get_entity(entity).last_changed_seconds / 60

void prawn
#

nice!

#

done

#

restarting AppDaemon

#

I don't see any errors

bitter skiff
#

just keep in mind til you fix that garage door open situation, no timers will execute for it

#

regular doors might work now though

void prawn
#

I'm going to go test the regular doors in a second. I do get notifications on the garage door from another automation in HA so that isn't a huge deal at the moment.

#

no notification for the regular doors either

bitter skiff
#

check the error log again

void prawn
#

2023-03-08 15:58:50.465396 WARNING door_notify: ------------------------------------------------------------
2023-03-08 15:58:50.466152 WARNING door_notify: Unexpected error in worker for App door_notify:
2023-03-08 15:58:50.466736 WARNING door_notify: Worker Ags: {'id': 'ade486f877c84b8098344214dea020b7', 'name': 'door_notify', 'objectid': '15f457708ee143bcbaca6f3a815e7c20', 'type': 'scheduler', 'function': <bound method DoorMonitor.notifier of <door_notify.DoorMonitor object at 0xffffafbaf100>>, 'pin_app': True, 'pin_thread': 8, 'kwargs': {'entity_name': 'binary_sensor.garage_entry_sensor', '__thread_id': 'thread-8'}}
2023-03-08 15:58:50.467281 WARNING door_notify: ------------------------------------------------------------
2023-03-08 15:58:50.468121 WARNING door_notify: Traceback (most recent call last):
File "/usr/lib/python3.10/site-packages/appdaemon/threading.py", line 904, in worker
funcref(self.AD.sched.sanitize_timer_kwargs(app, args["kwargs"]))
File "/config/appdaemon/apps/door_notifications/door_notify.py", line 47, in notifier
if elapsed_time > threshold:
TypeError: '>' not supported between instances of 'float' and 'datetime.timedelta'

2023-03-08 15:58:50.468713 WARNING door_notify: ------------------------------------------------------------

bitter skiff
#

Just use ttl as the comparison on line 47 @void prawn

#

You don't need a timedelta at that point

void prawn
#

so threshold = (minutes=ttl)??

#

this is my line 47 in the code: if elapsed_time > threshold:

#

here is lines 45 through 51
for ttl in cfg.get("ttl", [self.defaults.get("ttl", 1)]): threshold = datetime.timedelta(minutes=ttl) if elapsed_time > threshold: message = f"The {entity.split('.')[1].replace('_', ' ')} has been open for {elapsed_time}." self.log(f"Notifying {cfg.get('notify_service', self.defaults.get('notify_service', 'notify/notify'))}: {message}") self.call_service(cfg.get('notify_service', self.defaults.get('notify_service', 'notify/notify')), message=message) break

#

ohhhh . . . I just need to change elapsed_time > threshold to elapsed_time > ttl, correct?

#

@bitter skiff , I think that is correct. . . . so the next question is do I need to put ttl in quotes since it is referenced as a string above? so line 47 would read:
if elapsed_time > "ttl"

#

I did it without the quotes and have no errors so far. But I haven't opened a door yet.

bitter skiff
#

no, you don't want ttl in quotes there

void prawn
#

good deal! No errors, but no notifications

bitter skiff
#

new errors?

void prawn
#

0 errors

#

nothing in the appdaemon log and nothing in the error log

#

how can I add a debugging log to the notifier function?

#

here is the log for the door_noitfy file:
2023-03-08 16:39:49.041726 INFO door_notify: Entities: {'binary_sensor.garage_entry_sensor': {'ttl': [0.125, 2, 5, 10]}, 'binary_sensor.front_door_sensor': {'ttl': [0.125, 5, 10]}, 'cover.garage_door': {'ttl': [0.125, 2, 5, 10, 15, 20]}, 'binary_sensor.upstrs_back_door_sensor': {'ttl': [0.125, 5, 10]}}
2023-03-08 16:39:49.043040 INFO door_notify: Defaults: {'ttl': 0.125, 'notify_service': 'notify/notify'}
2023-03-08 16:39:49.044072 INFO door_notify: Notify Notify Entities: []
2023-03-08 16:46:26.937929 INFO door_notify: binary_sensor.garage_entry_sensor has been opened.
2023-03-08 16:46:33.978524 INFO door_notify: binary_sensor.garage_entry_sensor has been closed.

bitter skiff
#

self.log(f"In {entity} notifier") after your first line in there

void prawn
#

after the entity = kwards['entity_name'] ?

bitter skiff
#

yeah

void prawn
#

kwargs. . . not kwards

bitter skiff
#

probably also want to output the ttl and the elapsed_time

#

self.log(f"TTL: {ttl} - Elapsed Time: {elapsed_time}") before if elapsed_time > ttl

#

Those are probably equal actually and that's why it's not doing anything

#

At this point right now you're calling notifier once exactly at the TTL of the first value in the list

void prawn
#

I just added the recent language and havne't tested yet, but previously adding the notifier gave us this log output:

2023-03-08 16:55:33.007335 INFO door_notify: binary_sensor.garage_entry_sensor has been opened.
2023-03-08 16:55:33.017537 INFO door_notify: In binary_sensor.garage_entry_sensor notifier
2023-03-08 16:55:38.167024 INFO door_notify: binary_sensor.garage_entry_sensor has been closed.

#

I feel like it isn't using the notify/notify function (may not be the correct terminology)

bitter skiff
#

Well, you've got a log in there, so it'd either error or log if it made it inside that branch

#

So fairly confident in saying the elapsed_time <= ttl at this point

#

(likely = )

void prawn
#

recent log:
2023-03-08 16:59:09.810937 INFO door_notify: binary_sensor.garage_entry_sensor has been opened.
2023-03-08 16:59:10.005861 INFO door_notify: In binary_sensor.garage_entry_sensor notifier
2023-03-08 16:59:10.008897 INFO door_notify: TTL: 0.125 - Elapsed Time: 0.0035032166666666667
2023-03-08 16:59:10.010353 INFO door_notify: TTL: 2 - Elapsed Time: 0.0035032166666666667
2023-03-08 16:59:10.011692 INFO door_notify: TTL: 5 - Elapsed Time: 0.0035032166666666667
2023-03-08 16:59:10.012764 INFO door_notify: TTL: 10 - Elapsed Time: 0.0035032166666666667
2023-03-08 16:59:15.691672 INFO door_notify: binary_sensor.garage_entry_sensor has been closed.

#

so change if elapsed_time > ttl: TO if elapsed_time >= ttl:??

bitter skiff
#

Guess what int(ttl_list[0])*60 is when ttl_list[0] = 0.125? It's 0

#

So, notifier is being called immediately

void prawn
#

Should I still get a notificaiton then?

#

or is it calling notifier before ttl has elapsed and then breaking out and not waiting for the correct ttl?

#

do I need to delete the 0 within the brackets?

#

That isn't it. . .that broke it. . .haha!

#

do I need to enter the number of entities I will have in the array? In this case it is 4, so do I need to input a 4 in the brackets so it is looking for 4 entries?

#

@bitter skiff , I also have this twice in the code. . . .is that correct?? - see below

self.timers[entity] = self.run_in(self.notifier, int(ttl_list[4])*60, entity_name=entity)

void prawn
#

@bitter skiff I am getting some notifications for the front door and the back door. The garage entry is not notifying me.

bitter skiff
#

Those all appear to be set up the same way

#

except I suppose your garage entry has 4 ttls instead of 3

elder schooner
#

My add-on changelogs and documentations aren't loaded anymore since Supervisor update 2023.03.1 . Is it known and/or does anyone else have the same issue?

void prawn
void prawn
#

@bitter skiff whats your coffee link?

astral tiger
#

Do I need to do something in HA to make this visible? Seems to be part of the community pack so should be default

bitter skiff
bitter skiff
#

Was never technically released

astral tiger
#

Awesome thanks!

#

Wil add it manually and try

void prawn
#

@bitter skiff How do I add a round command to round the minutes to 4 decimal places?

bitter skiff
#

should just be able to round(the_variable_name, 4)

ivory geyser
#

Hello, I’ve been trying to get AdGuard Home to work and I have confirmed that changing the DNS on my Windows PC works, however inserting the IP of my HA yellow device into my router doesn’t

#

I can only connect to sites using direct IP

#

Just seems like really weird behavior and the logs doesn’t exactly have anything useful…

ivory geyser
#

Update: Got it to mostly work with DHCP editing, I think I was accidentally changing the connection modem's DNS....now it seems that my iphone won't send DNS info through AdGuard Home unless private network is off

charred spindle
#

Sounds like you're doing it all the hard way. Can't you just say your router's LAN DNS to the IP of your Yellow?

Keep in mind you may need to release and renew (or reboot) your devices to get the new DNS from the router

ivory geyser
#

It’s been a few hours now, I forced all wifi devices off the internet by turning off DHCP, disabling the wifi entirely, and then turning it back on. That seems like it forced a release and renew for all devices

merry zinc
ivory geyser
#

It seems like I do in fact get some requests from the phone from ATT and and Apple but that’s it

ivory geyser
merry zinc
#

Or if you want to set it at a network level, you would do the same. Manually set the DNS of the router to the IP:Port of the device hosting your AdGuard insstance

ivory geyser
#

I am able to change the DNS, and it does work for a few seconds before the router giving up and saying there’s no internet

merry zinc
#

because its not resolving, it's not making a connection to the DNS, and probably no fallback DNS is being provided

#

Did you use the correct port? are you using AdGuard in a Container? did you configure the container correctly for port access?

ivory geyser
#

I am unable to set a port for my DNS server

#

Default port is set to 53, and I’m running within a Home Assistant Yellow

#

Like I said before, seems I got it working now with everything using that DHCP server on my router. Just confused as to why my phone seems to bypass the DNS server altogether, even when it’s the only dns server listed in the network configuration

merry zinc
#

Well I was trying to post a screenshot of what setting needs to be configured, but for some reason you can't post photos in here.. or at least I can't 🤷🏼‍♂️

#

thats the DNS that should be configured, at the DHCP level.

ivory geyser
#

This is what I see, and I just put in the ip of the home Assistant device into the primary dns

merry zinc
#

Right, that's where it's suppose to go..

#

But you should set a Secondary, incase the Primary Fails

#

(the Yellow crashes and burns or whatever)

ivory geyser
#

Yeah I probably should :v

#

Just weird weird behavior

merry zinc
#

As for the iPhone. Power cycle it. so it reconnects to the network.

ivory geyser
#

Yeah, no difference

covert bobcat
#

Hello, i am a student and trying to make an add on that runs a server.js with node to start a json server but when i start the add on i cant reach the json server.. Can someone help me out?

supple grove
#

Share your code/config

void prawn
bitter skiff
#

That should just take place of wherever you want the rounded value

#

replacing the_variable_name, of course

void prawn
#

Here is the code I think this applies to. I believe it is the elapsed time I want to round. I'm honestly don't know where to insert thr round command

elapsed_time = self.time_since_changed(entity) for ttl in cfg.get("ttl", [self.defaults.get("ttl", 1)]): threshold = datetime.timedelta(minutes=ttl) self.log(f"TTL: {ttl} - Elapsed Time: {elapsed_time}") if elapsed_time > ttl: message = f"The {entity.split('.')[1].replace('_', ' ')} has been open for {elapsed_time}."

#

forgive the spelling. I'm typing with one hand while I hold the baby

#

Would this do it?

elapsed_time = self.time_since_changed(entity) for ttl in cfg.get("ttl", [self.defaults.get("ttl", 1)]): threshold = datetime.timedelta(minutes=ttl) self.log(f"TTL: {ttl} - Elapsed Time: {elapsed_time}") if elapsed_time > ttl: round({elapsed_time}, 4) message = f"The {entity.split('.')[1].replace('_', ' ')} has been open for {elapsed_time}."

void prawn
#

This code is acting strange. The garage_entry entity doesn't notify it's opening every time (it is sporadic and there's not pattern). I also only get one notification per door opening and not after each ttl. any thoughts on that?

viscid orchid
light pecan
elder schooner
light pecan
ivory geyser
#

I just want to add I have figured out what was going on with my iPhone not using the DNS settings I had set. Turns out I had this tweak called MYbloXX, which was labeled to be a system wide ad blocker. Didn't know how it worked and it turns out it was overriding my DNS server that my router was giving out as a way to block ads.

#

Want to add this in case anyone else was having anything similar, tho I might be literally the only person in the world to have this funky configuration.

#

I did in fact do everything right, I just had other programs conflicting with each other

barren moat
#

Hi Guys, Is there any way to create a switch in HA which will turn on and off a tunnel for my locally hosted service using Cloudflared Add on ?

bitter skiff
versed summit
#

Hi Everyone, Looking for some guidance to add my alarm system to HA. I have used the alarm decoder ad2serial with Vera. Not sure why but can get it to work with HA. Any help would be appreciated.

barren moat
ancient beacon
#

Is there a way to restart addon throught CLI. I want to restart them once I upload my new SSL certs for the addon

bitter skiff
#

Otherwise ha addons restart addon_slug

uneven spade
#

I can use camera.snapshot from esp32 cam, but not from my other #cameras-archived which are using WebRTC ?
Could someone enlighten me on this?

ancient beacon
#

Service call won't work since I'm sending the command through SSH using HA I get an error. I used "docker restart addon_addon_slug

barren moat
#

you can get a list of add-on with ha addons list command

pine narwhal
#

Is there any way to renew duckdns certificate?

#

This is the log from the DuckDNS add-on

#
  • Checking domain name(s) of existing cert... unchanged.
  • Checking expire date of existing cert...
  • Valid till May 8 18:27:41 2023 GMT (Longer than 30 days). Skipping renew!
    [23:35:15] INFO: OK
    77.105.54.68
    NOCHANGE
    [23:40:18] INFO: OK
    77.105.54.68
    NOCHANGE
sly kindle
#

It doesn't require renewal?

#

What is it you actually need

pine narwhal
#

How do you mean it doesn't require renewal?

#
  • Valid till May 8 18:27:41 2023 GMT
#

And my antivirus tells me that there is a problem with the certificate when I access through the browser. I can't access it through the app on my phone.

mental kiln
#

[Plex addon], how do you add shares to the file system to use as libraries and is it possible to use hardware transcoding when available? My system has the required hardware. Does this addon pick that up or do I need to configure something with it for the addon to use the gpu correctly?

tropic pilot
supple grove
void prawn
#

I am so close to having this working it isn't even funny. I have had notifications working but the loop wasn't working correctly for each ttl. . . now I'm not getting any notifications but the log is working correctly right up until the notifier function. Any thoughts on how I can get it to notify me as soon as ttl exceeds the ttl set in my yaml configuration? Python code below:

https://pastebin.com/FcS0A5Aw

yaml config:
door_notify:
module: door_notify
class: DoorMonitor
log: door_notify

Common configuration for all entities

defaults:
ttl: 0.125
notify_service: notify/notify

Entity-specific configuration

entities:
binary_sensor.garage_entry_sensor:
ttl: [0.00005, 1, 10]

binary_sensor.front_door_sensor:
  ttl: [0.003, 5, 10]

cover.garage_door:
  ttl: [0.003, 2, 5, 10, 15, 20]

binary_sensor.upstrs_back_door_sensor:
  ttl: [0.003, 5, 10]
viscid orchid
#

AppDaemon isn't an official addon

void prawn
#

Oh nice!!! Thank you!

fringe iris
#

I need help, I installed home assistant on my windows 10 pc today and it is working fine until I attempt to install hacs, if you would like to help me, then please dm me. Thank you for your consideration

limpid canyon
#

Just accessed Glances on HA only to see addon_core_silabs_multiprotocol pegged at 20% CPU. Seems V high to me for that. I know there's a lot of work being done on this, anyone else spotted this?

tough breach
#

Hey does anyone know how to setup nanoleaf lightstrips via thread if so can you help me?

modest heath
#

hi, is there someone who could help with wireguard and HA? I already have Wireguard on my VPS on the internet and i just want to somehow connect local HA to make it reachable from outside network. Every "how to" are just manuals for wireguard server in HA (thats not what i want, i have already wireguard server in VPS). Thx

fringe iris
#

I need help, I installed home assistant on my windows 10 pc today and it is working fine until I attempt to install hacs, if you would like to help me, then please dm me. Thank you for your consideration

#

I need help, I installed home assistant on my windows 10 pc today and it is working fine until I attempt to install hacs, if you would like to help me, then please dm me. Thank you for your consideration

#

I need help, I installed home assistant on my windows 10 pc today and it is working fine until I attempt to install hacs, if you would like to help me, then please dm me. Thank you for your consideration

brave basin
#

I'm following a tutorial on setting up Grafana, and notice a difference I think happened with a new version. I'm looking for the options for visualization on the right, where it shows little thumbnails of different chart options. Is that still available somewhere?

bitter skiff
#

Click it

brave basin
#

Ahh thanks! And any idea where the "auto refresh" went?

bitter skiff
#

I see it when I'm viewing a dashboard

marsh pawn
#

i am having a strange issue where i am unable to browse the config folder through any add-on. despite this, home assistant boots and works fine, both vscode and file editor seem unable to locate the file structure

marsh pawn
#

from what i can tell the add ons need elevated access to access the config folder, not sure if that used to be the case, but it doesnt seem to have that access

sly kindle
#

@marsh pawn if file editor and vscode show the configuration.yaml file, that is the config folder

marsh pawn
#

No, neither do

#

I had to use the web terminal and enable elevated access to make changes to the config.yaml since the file browsers aren't working for some reason

patent magnet
#

Hi everyone,
after restoring a backup (from last night), my recorder, history etc. is no longer working.
According to the log, the database is not available.
I am using MariaDB and it is showing no error itself.

# 
s6-rc: info: service s6rc-oneshot-runner: starting
s6-rc: info: service s6rc-oneshot-runner successfully started
s6-rc: info: service fix-attrs: starting
s6-rc: info: service fix-attrs successfully started
s6-rc: info: service legacy-cont-init: starting
s6-rc: info: service legacy-cont-init successfully started
s6-rc: info: service legacy-services: starting
services-up: info: copying legacy longrun mariadb (no readiness notification)
services-up: info: copying legacy longrun mariadb-lock (no readiness notification)
[21:54:57] INFO: Using existing mariadb initial system
[21:54:57] INFO: Starting MariaDB```
#

Is this a known issue when restoring HA? Does anybody have a clue how I can fix it?

storm elk
#

hello, i'm using HAOS and trying to get the nginx ssl proxy add-on working, but the http integration is blocking the request. i see that i need to add a trusted proxy, but i can't find any way to edit configuration.yaml or the http integration to do so. all the instructions seem to be for if you installed home assistant on a linux box, not using HAOS. could someone help me understand what i'm doing wrong?

viscid orchid
storm elk
patent magnet
#

Does anybody know where the mariadb database is being stored? I uninstalled and reinstalled it and my database is gone. I would like to restore my backup of the database but cannot find the folder to copy to

supple grove
#

You should use the backup functionality of HA for generic backups and PhpMyAdmin or another client to export dumps.
The actual files are stored in /mnt/data/supervisor/addons/data/core_mariadb/ but good luck restoring them without breaking something

patent magnet
#

Problem is that I used the HA restore and IT broke the DB.

#

Folder /mnt/ seems to be empty

supple grove
#

HAOS is full of surprises 🙂
Restoring a broken database backup might be unfeasible. You need proper SSH access

patent magnet
#

I am running docker, but I do have SSH

supple grove
#

Huh? You are running HAOS, right?. The SSH addon is just a container that does not give you access to everything on the host

patent magnet
#

No, HA supervised

#

Btw, I cannot select to restore the MAriaDB database from BAckups

supple grove
#

I see. I'm not sure where the supervised version stores its data. Maybe in /var/lib/docker

patent magnet
#

no such folder either 😦

supple grove
#

You could try something like find / -name ibdata1 or just check the mounts for the MariaDB container

#

If all else fails you can just create a VM. Install MariaDB. Restore/repair the files there and then dump the data via a client and import it in HA 😓

wild basin
patent magnet
supple grove
#

You import .sql files, not idb

supple grove
wild basin
#

yes

wild basin
# supple grove Are you running HAOS?

Version core-2023.3.3
Type d'installation Home Assistant OS
Développement false
Supervisor true
Docker true
Utilisateur root
Environnement virtuel false
Version de Python 3.10.10
Famille du système d'exploitation Linux
Version du système d'exploitation 5.15.84-v8
Architecture du processeur aarch64
Fuseau horaire UTC
Répertoire de configuration /config

patent magnet
supple grove
patent magnet
#

So I am back to the issue of where those files need to be copied to as I cannot import them directly?

supple grove
supple grove
patent magnet
#

find found nothing (executed in HA terminal)

supple grove
#

Not sure what HA terminal means

patent magnet
#

The HA addon "SSH & Web Terminal"

supple grove
#

Wait a minute. Is MariaDB currently installed/running on your host?

patent magnet
#

Installed yes, but stopped

supple grove
#

I told you earlier that the SSH addon is not going to work for this. Only certain things are mounted into the container. Use proper SSH on the host

patent magnet
#

Where do I connect to then?
IP:8123?

wild basin
#

thanks @supple grove

supple grove
#

No. 8123 is the (default) port of HA. SSH usually runs on 22 but only you know how your server is set up

patent magnet
#

Problem is that raspberry pi itself has an ssh server running on 22

supple grove
#

Why is this a problem? I'm not following

#

You can probably also do something like this

docker inspect nameoridofmariadbcontainerhere -f '{{ .HostConfig.Binds }} - {{ .HostConfig.Mounts }}'
patent magnet
#

I connected via SSH but find does not find anything

supple grove
#

Then check docker ps for the container name and then the command above to see where it mounts its data

patent magnet
#

Template parsing error: template: :1:14: executing "" at <.HostConfig.Binds>: map has no entry for key "HostConfig"

supple grove
#

Try just '{{ .HostConfig.Mounts }}' then

wild basin
#

I want to install the tydom addon, but it asks me for the server's mqqt port and address, what is the way to do?

supple grove
#

Depends on your setup

patent magnet
#

Ah, cannot post a picture or Portainer here

wild basin
#

I don't have to use mqqt directly linked to tydom?

supple grove
#

No idea how this specific addon works. I don't use it

patent magnet
#

I got the ID wrong

#

Now I am in

wild basin
#

2023-03-12 00:41:55,391 - mqtt_client - INFO - MQTT error, restarting in 8s...
2023-03-12 00:42:03,403 - mqtt_client - INFO - """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
2023-03-12 00:42:03,405 - mqtt_client - INFO - Attempting MQTT connection...
2023-03-12 00:42:03,405 - mqtt_client - INFO - MQTT host : localhost
2023-03-12 00:42:03,406 - mqtt_client - INFO - MQTT user :
2023-03-12 00:42:03,415 - mqtt_client - INFO - MQTT connection Error : Multiple exceptions: [Errno 111] Connect call failed ('127.0.0.1', 1883), [Errno 99] Address not available
2023-03-12 00:42:03,415 - mqtt_client - INFO - MQTT error, restarting in 8s...

#

nice..

patent magnet
# 
pi@raspberrypi:~ $ sudo docker inspect 5d4f53b1bb4b757e25116a8acf20b986037d57fd55d0de8e98b68862c02d64ad -f '{{ .HostConfig.Binds }} - {{ .HostConfig.Mounts }}'
[/dev:/dev:ro /usr/share/hassio/addons/data/core_mariadb:/data:rw] - []
pi@raspberrypi:~ $
supple grove
#

Do a ls -lhR /usr/share/hassio/addons/data/core_mariadb

patent magnet
#

That looks promising

# 
/usr/share/hassio/addons/data/core_mariadb:
total 12K
drwx------ 7 root root 4.0K Mar 12 00:33 databases
-rw------- 1 root root  243 Mar 12 00:33 options.json
-rw-r--r-- 1 root root   65 Mar 11 22:37 secret
ls: cannot open directory '/usr/share/hassio/addons/data/core_mariadb/databases': Permission denied
pi@raspberrypi:~ $
supple grove
#

I suggest you back up core_mariadb or just databases, make sure the container is stopped and then move the contents of your MariaDB backup in the folder that looks like it belongs. Make sure you restore the permissions or better, work as root for this.
The files should look similar (ibdata1 and so on) so it's hard to miss where to put them

#

You can then start the addon again and follow the logs with docker logs -f nameofcontainerhere

#

I'm surprised find didn't find that folder. There should be a ibdata1 in there

supple grove
wild basin
#

Ok I install it thanks

patent magnet
supple grove
#

I'm not that hopeful to be honest but let's see

patent magnet
#

Hmmm. Restarted HA and it is not coming back up. Either it is trying to repair the database and that needs time, or HA is not booting.

patent magnet
#

Looking good now. Reboot RPi and after a few seconds, HA booted and the data seems to be there. Actually seems like I have the old database plus the data since the backup. Need to check more tomorrow 🙂

#

Thank you very much @supple grove 🫵👏👏

real seal
#

hi, i have a lot of auto detected rtl433 entities. i.e. tire sensors.

#

how may i delete them?

#

they are shown as not available

supple grove
brave basin
#

Does anyone know how I can "invert" the thresholds in a grafana gauge? I want the "red" to be below a number, not above.

#

Is it just a matter of reversing the colours?

supple grove
#

Yep

tropic pilot
unkempt flame
bitter skiff
#

That's not an add-on, it's an integration

#

Do you have HACS installed?

#

That'll be the easiest way to get it

supple grove
weary stratus
#

hallo I had vlc-telnet happily working then at some point I cannot use it anymore to play files, if I try to add it again I am asked for a password, lost there, been fiddling with the documetnation but cannot get it back on track

tropic pilot
weary stratus
#

the VLC addon is working properly, but I am not sure about the password it is asking for

supple grove
#

Addon is a HAOS term so that's confusing. You can clone the repo and build the Dockerfile yourself

tropic pilot
supple grove
#

It's pretty much just docker build . inside the directory where Dockerfile is
For compose it's build: .

tropic pilot
#

I’m new to docker & compose. No idea where to start.

supple grove
mortal moth
#

I am starting my first addon (after the hello_world tutorial). It will use PHP. I am thinking that using PHPs built in web server may be a simple lightweight way of getting up and running.

#

In my Dockerfile I have RUN apk add --no-cache php8 php8-common

#

In run.sh I have php -S localhost:8000

#

In the add-on log I see [Sun Mar 12 18:09:45 2023] PHP 8.0.28 Development Server (http://localhost:8000) started but no further entries after this.

#

If I do curl http://localhost:8000 from my client (which is running the devcontainer) I see curl: (52) Empty reply from server

supple grove
#

Try with 0.0.0.0

mortal moth
#

Same thing with php -S 0.0.0.0:8000

mortal moth
#

Ok - there seems to be an issue with the specific php8 package. If I just add php then it works.

visual jewel
#

was wondering if there's anyone around knowledgeable with the RTLAMR2MQTT addon?

#

or generally ... how to figure out the meter ID of my utility meters

#

by looking at the nameplates on my meters, nothing really stands out as what I think is the meter ID; when looking at my monthly bill, there is a number on it that matches something on the meter ... it's a 6 digit number and so I think there's a good chance that it is the meter ID. However, I can't figure out what is the protocol for it

full rune
#

ZwaveJS2MQTT

quiet apex
#

Hello, I am using the EMQX MQTT addon as my broker and have a local and remote zigbee2mqtt instance running. When I restart home assistant, my mqtt devices are all unavailable from both instances. If I restart both Z2M instances, everything comes back up. Searching shows this is related to the birth and lwa messages. I looked at the MQTT integration configuration and auto discovery, birth and will messages are all enabled. Any idea what I'm doing wrong?

viral fractal
# viral fractal My research so far indicates that a HM-MOD-RPI-PCB connected to J11 should work ...

I got my HM-MOD-RPI-PCB working now with Yellow and the RaspberryMatic Add-On! I needed to edit the config.txt which is on a partition usually not available during normal operation. I was able to edit this via the USB-UART interface following this guide: https://yellow.home-assistant.io/guides/use-serial-console-linux-macos/
Once I got access to the config txt, I followed this guide https://github.com/jens-maus/RaspberryMatic/wiki/Installation-HomeAssistant , especially section "Using HomeAssistant OS" where I made sure the following lines in config.txt are uncommented:
enable_uart=1 dtparam=i2c_arm=on dtoverlay=miniuart-bt dtoverlay=rpi-rf-mod
After adding the homematic integration via yaml and using the RaspberryMatic Addon-Hostname as the host I was able to access my homematic entities!

viral fractal
#

Was my config.txt changes too much? Would it have been enouth to only uncomment the dtoverlay=rpi_rf_mod option?

#

Or is it impossible to use J11 in conjunction with the integrated Zigbee module?

weary stratus
#

Hi, I receive always 502: Bad Gateway node red, re-installed it several times.

#

This is my config

#

certfile: fullchain.pem
credential_secret: hereIaddedmysecret
http_node:
password: ""
username: ""
http_static:
password: ""
username: ""
init_commands: []
keyfile: privkey.pem
npm_packages: []
ssl: false
system_packages: []
theme: default

bitter skiff
#

sounds like you have to look at your log

old verge
#

I have ssh installed, it shows running, but I am getting "Connection Refused" when I attempt to ssh locally

viscid orchid
#

did you specify a port?

quick trench
#

(in the addon)

old verge
#

yessir. 22

viscid orchid
#

and give it a password or key? The logs will tell you what's going on

old verge
#

No key, but yes a pw

viscid orchid
#

so...logs

quick trench
#

also, did you acutally type out 22 in the SSH Port or did you leave it blank because it had a grey 22 in it already?

viscid orchid
#

the default is to only enable the web terminal

old verge
#

used the up arrow to set the port, clicked save, then restarted the add-on

quick trench
#

Then it should be working, check the log for details on why it's rejecting the connection

viscid orchid
#

if you have a terrible password, for instance

quick trench
#

Oh? Does this powned too?

viscid orchid
#

I think so?

quick trench
#

interesting

#

I always use the sha2 key

viscid orchid
#

maybe not, I don't know, and the docs don't explicitly say. I also use keys

#

I'm using a container install now and can't check

fresh moth
#

what does it say under the Log tab?

old verge
#

we aren't even getting to a pw request

#

just restarted for the umpteenth time... now it's saying FATAL: You need to setup a login!

#

that's new

#

grumbles

viscid orchid
#

Probably the exact same problem it's always been

old verge
#

still nojoy. thanks, but I give up. this thing is still in alpha testing imho

fresh moth
#

pretty stable to me, but you do you

viscid orchid
#

adding more containers isn't supported

#

the portainer addon was removed a while ago

supple grove
#

Seems like it looks for the existence of watchtower, ouroboros, and portainer
There's a third party repo that provides portainer as addon if you absolutely have to have it

rose siren
#

Has anyone found a solution to “s6-overlay-suexec: fatal: can only run as pid 1” with the ring live streaming add on. I am running HAOS on a re-commissioned old PC not via docker, or similar. Thanks loads to anyone who can point me in a general direction. Every search I’ve done ends up with a docker or portainer solution.

supple grove
#

You should post the full log, not just one line and maybe a link to the addon

viscid orchid
#

that just tells me that it either hasn't been updated in a while, or you're not using the latest version

wind marsh
#

Can anyone help me here, I receive that message when trying to install my first add-on:

Can’t install homeassistant/aarch64-addon-duckdns:1.15.0: 500 Server Error for http+docker://localhost/v1.41/images/create?tag=1.15.0&fromImage=homeassistant%2Faarch64-addon-duckdns&platform=linux%2Farm64: Internal Server Error (“Get “https://registry-1.docker.io/v2/”: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)”)
If anyone feel helpful, here is the link to the thread where I provide more details:
https://community.home-assistant.io/t/problem-with-dns-configuration/547783/4

brave basin
#

I've been having problems with my HAOS crashing whenever I access my VSCode remotely (using cloudflare tunnel). I use https, so that can't be the issue, but does anyone have any tips on how to troubleshoot?

supple grove
zinc mortar
zinc mortar
wind marsh
wind marsh
wild zephyr
#

I couldn't find on the addons dot community, does anyone know of an addon that adds Apple MIDI to Home Assistant? tysm!

full shuttle
#

I'm trying to rebuild an addon, and when I click Rebuild in the UI, it just gives me an exclamation mark but no error. There are no messages in the log files. What is going on?

#

If I navigate to the API endpoint it's trying to hit, the page doesn't return anything.

viscid orchid
copper cove
#

Hi all, I'm having issues with the HA add-on NGINX. It seems not to start
[14:12:22] INFO: Generating dhparams (this will take some time)...
Generating DSA parameters, 4096 bit long prime

Can't figure out what is wrong here. DuckDNS is installed and NGINX was required to use Reolink doorbell...
Networking is not my cup of tea, so any help would be much appreciated

sly kindle
#

Have you clicked on "refresh" on the logs page?

copper cove
#

yes.... cleared browser cache and more.... but it fails to start....

jade bay
#

Hi all, getting a nextcloud error that it can't find the app directory. I was wondering if anyone knows where the config files are that I can completely wipe clean on HAOS? I assumed they'd be in /config/addons_config but I can't see the nextcloud file in there. Thanks

teal crater
#

I just installed and configured the Samba share addon. I hit start and it told me to configure a password first. I ddid that and hit start again and nothing happened. I can connect to the share now, but the addon still has the option to start (vs stop), meaning I cannot stop it. I'm not sure if it will persist this way after a reboot yet, but are there any known scenarios that might lead to this, or should it potentially be reported as a bug? Since all I did is configure the password, I'm thinking maybe my password has a character something doesn't like?

ETA: I later installed the AppDaemon add-on and it did the same thing. After that, I migrated from a RPi 3b to a Yellow and all add-ons did that. A couple reboots (not restarts) later (with tinkering in between), it was all working.

restive lily
#

anyone use wmbusmeters addon with a nooelec rtl sdr? I'm trying to figure out the path to my usb device.

bitter skiff
#

Use the /dev/serial/by-id/... path

frank plover
#

Hi. If I already have a domain registered (mine is with 123-reg.net) how can I employ that in an attempt to secure external access to my Home assistant instance? ...or must I create a new domain with DuckDNS? Thanks.

astral whale
#

Is there any way to add an additional port forward in the official NGINX add-on?

remote marlin
#

is anyone of you actually using hass.agent with zigbee2mqtt? it just doesnt connect to my instance, aren't they compatible?

bitter skiff
bitter skiff
remote marlin
#

sorry, confused names, obviously meant mosquitto

bitter skiff
#

Then yes, they are compatible

remote marlin
#

when looking into the mosquitto logs, i cant see the request pop up there

bitter skiff
#

You perhaps haven't exposed the MQTT port of the addon

remote marlin
#

disabling the firewall in windows doesnt change

#

how to do so?

bitter skiff
#

Configuration

#

Ensure 1883 is filled in the Network section for Normal MQTT

remote marlin
#

hm, am using core tho, but the port should be filled properly, at least my device tab says

bitter skiff
#

As in... you're not using an addon

remote marlin
bitter skiff
#

I'm not sure what I'm looking at

remote marlin
#

i just hope thats the same representation that you would find in supervisor? basically the configure screen of mqtt in the HA device overview

bitter skiff
#

How are you hosting Mosquitto

#

Just as a docker container not part of HA OS?

#

You need to expose the Mosquitto port 1883 from the container through the host

remote marlin
#

not even that, manually, no container

bitter skiff
#

Then UFW or whatever firewall on the box itself?

remote marlin
#

how would i find out? its a raspberry pi

# 
cornstar@himbeerkuchen:~ $ sudo lsof -i:1883
COMMAND       PID          USER   FD   TYPE   DEVICE SIZE/OFF NODE NAME
mosquitto 1082634     mosquitto    5u  IPv4  xxxxxxx      0t0  TCP localhost:1883 (LISTEN)
mosquitto 1082634     mosquitto    6u  IPv6  xxxxxxx      0t0  TCP localhost:1883 (LISTEN)
mosquitto 1082634     mosquitto    8u  IPv4 xxxxxxxx      0t0  TCP localhost:1883->localhost:44498 (ESTABLISHED)
mosquitto 1082634     mosquitto    9u  IPv6 xxxxxxxx      0t0  TCP localhost:1883->localhost:46199 (ESTABLISHED)
hass      2818070 homeassistant   48u  IPv6 xxxxxxxx      0t0  TCP localhost:46199->localhost:1883 (ESTABLISHED)
node      3761356      cornstar   20u  IPv4 xxxxxxxx      0t0  TCP localhost:44498->localhost:1883 (ESTABLISHED)
#

so locally the port is fine it seems, right?

bitter skiff
#

Potentially only listening on lo?

#

What's your listener configured for in Mosquitto

remote marlin
#

like its conf?

bitter skiff
#

Correct

remote marlin
#

i cant remember setting much there....

#

let me find it

#

but when starting it, it says

#

1679087024: Starting in local only mode. Connections will only be possible from clients running on this machine.

#

so apparently thats our reason

bitter skiff
#

yup

remote marlin
#

so i add listener 1883 to the conf i guess?

bitter skiff
#

well, is bind_host or bind_interface set

#

you'll probably have to specify the local IP after the port

remote marlin
#

in the comments it says that bind thingy too

#

listener 1883
bind_interface eth0

#

at least a telnet towards that port doesnt state an error, but still mqtt doesnt connect

#

and a allow_anonymous true did the trick then

#

thanks for the heads-up!

#

but apparently i just broke zigbee2mqtt with that -.-

# 
cornstar@himbeerkuchen:/etc/mosquitto/conf.d $ sudo systemctl status zigbee2mqtt.service
● zigbee2mqtt.service - zigbee2mqtt
     Loaded: loaded (/etc/systemd/system/zigbee2mqtt.service; enabled; vendor preset: enabled)
     Active: activating (auto-restart) (Result: exit-code) since Fri 2023-03-17 22:28:51 CET; 2s ago
    Process: 2948744 ExecStart=/usr/bin/npm start (code=exited, status=1/FAILURE)
# 
error 2023-03-17 22:31:10: MQTT error: connect ECONNREFUSED 127.0.0.1:1883
error 2023-03-17 22:31:10: MQTT failed to connect, exiting...
#

apparently worked without the bind_interface

weary stratus
#

Hi anyone can help with add users to ftp addon in HA. I'm adding in yaml and nothing works

river wharfBOT
#

@weary stratus I converted your message into a file since it's above 15 lines :+1:

📎 message.txt
weary stratus
#

Athentication faild for "hassio"

weary stratus
#

Sorry it's a Samba

tender ridge
#

Hoi, is there an addon for HA to use a hours watched, as well as how many episodes watched, like how i can see a log of prices and usage for power for X amount of time like a graph for plex?

kindred timber
#

Anyone know of signal-k and if an addon port is in the works? Has it been discussed anywhere?

oak ledge
#

][D][api:102]: Accepted 192.168.0.130
[10:46:34][W][api.connection:071]: 192.168.0.130: Socket operation failed: HANDSHAKESTATE_READ_FAILED errno=11

#

Help

#

What should I do?

#

ESP 32 error

oak ledge
#

I solved the problem. I was using space in the name of the ESP 32

north vault
#

Is it possible to backup my addons using the google drive add on? What about the data on the MariaDB?

silk root
#

how to backup to mega.nz using rc clone?

lucid terrace
#

how can I set a fixed IP on the nginx addon? There no longer is any relevant input on the "Network" section of the configuration

#

... or is there another way to use nginx in "bridged" mode so as it does not require it's IP to be cleared on the trusted_proxies: setting on the configuration.yaml file?

bitter skiff
#

You can just trust the docker network

# 
  trusted_proxies:
    - 172.30.33.0/24
lucid terrace
#

the logs of the "NGINX Home Assistant SSL proxy" are also not helpful at all, will not log / show failed connections or provide any useful debug information

velvet socket
restive lily
#

anyone here using WMBusMeters with a Nooelec RTL-SDR?

latent iron
bitter skiff
#

Nope, not sure

latent iron
bitter skiff
void urchin
#

Does anyone have a grafana panel to Monitor the uptime of shellys? I Do Not get it working... Which fiel Do i need to use? 🤐

supple grove
#

Where are you storing the uptime? Grafana has to query something

void urchin
#

I have the influxdb and MySQL as data source... So i can choose state and Filter the uptime entities.. But i dont get the right fiels for any aggregation...

supple grove
#

How does your query look right now?
Try something like this

SELECT mean("value") FROM "s" WHERE ("entity_id" = 'yourshelly') AND $timeFilter GROUP BY time($__interval) fill(none)
void urchin
#

Cant find any entity (except vaccum) within the option 's'. Within 'state' i have the uptime entities but there is 'No Data'...

supple grove
#

Can you share how it looks in the developer tools' states tab?

blazing laurel
#

Hey guys!

Dont know if this is the right subforum, or even the right discord but here goes:
Im running HA on a VM installed on a Lubuntu system. Its running fine, i can access it from local network fine.
My router is an ASUS RT-AC1200 V2 with OPENWRT OpenWrt 22.03.3.

Im trying to get remote access to my homeassistant with a duckdns.org domain, using ddns-scrpits installed on my router.
Im not getting anywhere. I tried tips from a combination of forumposts and guides, and im starting to feel like a complete idiot.

If anyone wanna give it a go i´d appreciate it!

supple grove
blazing laurel
#

Ill try that, Thanks!

brittle night
supple grove
#

Off-topic but also look into Tailscale

blazing laurel
supple grove
#

I respect that

brittle night
#

duckdns has been failing multiple times in 2023 alone. you can search "duckdns" here and see all the issues

#

but feel free to enjoy the pain

blazing laurel
#

well third times the charm, if that´s the case i might as well accept defeat.

sly kindle
#

As long as you don't plan on using it to stream media

flint plank
#

Good
Can someone help me my zigbee2mqtt is not working properly as the devices are working but it does not open the webpage.
Does anyone know what I do??

sly kindle
#

Not that they can't, but likely don't want to be dealing with plex servers etc. Against ToS

#

But yeah everyone creating a free CF account to start their own streaming service is not going to be something they are cool with

compact terrace
#

So I have come across the browserless add on. It’s essentially some headless version of chromium running on port 3000 with puppeteer. It seems I can use it with a screenshot API call to take a screenshot of any website and in particular: my HA dashboard to use on Esphome. Problem: I can’t figure how to log into HA, submit the form, etc. You can use curl to post a bunch of data for puppeteer, but I don’t know how to get it to fill the login forms. Does anyone have experience?

#

This is what I have curl -X POST http://ha.lan:3000/screenshot --output bla.jpg -H 'Cache-Control: no-cache' -H 'Content-Type: application/json' -d '{"url": "http://ha.lan:8123/lovelace/default_view", "options": { "fullPage": true, "type": "jpeg", "quality": 75}, "gotoOptions": {"waitUntil": "networkidle2"}, "authenticate": {"username": "blabla”, "password": "blabla"} }'

#

I’m guessing the authenticate only works with simple auth, but I am waaay outside my expertise

errant onyx
#

has anyone successfully used the mosquitto broker to broker to bridge to adafruit? I can't receive data and I keep getting connected/disconnected every 5 seconds or so

errant onyx
supple grove
spice bane
#

hey guys. does someone use esp-home? i accidentally installed nginx next to home assistant on my server and now my webinterface from esp-home is gone but all installed devices are still working. i tryed to remove nginx but the broblem was not solved. its not a big deal for me to reinstall the whole system but i want to get the scripts from my esp devices. is there a way to get them saved before? or maybe someone knows how to fix the main problem.

supple grove
spice bane
#

here is my current log. i restarted the docker container and the whole computer couple times

#

looks like some problems with ssl but i never changed a port or even enabled ssl before

#

i think i got it to work when i totally disable my firewall with sudo ufw disable but i dont know which ports i have to open for esp-home

supple grove
#

I meant the container's logs. Are you using supervised install?

spice bane
#

yeah i installed supervised

supple grove
spice bane
#

ah thanks for the idea just to check the firewall log file was great xDDD i got it back to work. port 65086 needs to be open aswell

supple grove
#

I'd check if that port changes if the addon/host restarts. Not too familiar with how addons talk to HA

flint plank
#

Good
Can someone help me my zigbee2mqtt is not working properly as the devices are working but it does not open the web page.
Does anyone know what I do??

wise thunder
#

Hi! Is there a way to add more local ip addresses to haOS? I have the NPM (nginx proxy manager) running on port 80 & 443. But i would also like to use the guest portal from Unify contoller i have running, currently runs at 8843, that i would like to have running on 443 for a nicely URL where my guests get redirected to. Im also interested in other solutions 🙂

#

The problem is that the redirect url of the unify controller now is https://<url>:8843/stuffandthings. I can change the port to 443 if i dont use NPM. But then i wont have a ssl cert.

hollow temple
#

Has someone an Idea, what I'm doing wrong?

#

[14:36:00] INFO: Preparing to start... [14:36:00] INFO: Socat not enabled [14:36:00] INFO: Starting Zigbee2MQTT... Zigbee2MQTT:info 2023-03-22 14:36:01: Logging to console and directory: '/config/zigbee2mqtt/log/2023-03-22.14-36-01' filename: log.txt Zigbee2MQTT:info 2023-03-22 14:36:01: Starting Zigbee2MQTT version 1.30.2 (commit #unknown) Zigbee2MQTT:info 2023-03-22 14:36:01: Starting zigbee-herdsman (0.14.96) Error: Reset error: Error: {"sequence":-1} after 2000ms at /app/node_modules/zigbee-herdsman/src/adapter/ezsp/driver/uart.ts:298:23 at Queue.executeNext (/app/node_modules/zigbee-herdsman/src/utils/queue.ts:32:32)

#

under "serial" i got that config here:

#

port: >- /dev/serial/by-id/usb-Nabu_Casa_SkyConnect_v1.0_2020d0168718ec11997be99a47486eb0-if00-port0 adapter: ezsp

brittle night
hollow temple
errant onyx
errant onyx
wise thunder
#

Yea it does. But i cant just add a hostname it has to include a port. And as unify also adds a port. I got like a double port, if you understand what im saying. So i thought i change the https portal of unify to 443 and NAT the nginx proxy from 443 to 4331. So it would work from the outside.

errant onyx
#

nginx should be able to just handle it. you'd point all the (sub)domains to the server. nginx listens on 80/443 and then redirects to the other addresses/ports internally

marsh pawn
#

im having issues with zigbee on my skyconnect specifically when open thread border router is enabled in the silicon labs multiprotocol add on

mortal panther
#

Does anyone here have an issue where studio code server does not run on android mobile app?

wise thunder
lunar ore
#

@raven tinsel could you please have a look at the tailscale addon? It hasn't been updated and is currently vulnerable to CVE-2022-41925. There have been multiple pull requests open for months. We just need you to approve any of them. A lot of us are getting desperate 🙂
https://github.com/hassio-addons/addon-tailscale/pull/133

raven tinsel
#

I get to it when I get to it

thick furnace
#

Hi all. I have little bit problem with InfluxDB. I have entities from my Goodwe Integration, in Homeassistant is all fine, i see changes in entities, in development tools is all fine, when i manually change state of entity, after few sec changes back to the right one. But from 3days ago, i have no record in Influx from this entity, from another entities is all fine. Only 2 entities dont send data to Influx battery_mode_code and interter_mode_code, can anybody help me with this pleas?

lunar ore
#

@raven tinsel thank you for your reply. Looking forward to when you get to it!

brittle night
cursive tusk
brittle night
#

@lunar ore i see you asked the guy submitting the PR how to use his fork. it's pretty simple and outlined in his docs. "Navigate in your Home Assistant frontend to Settings -> Add-ons -> Add-on Store, in the ... menu at the top right corner click Repositories, add https://github.com/lmagyar/homeassistant-addon-tailscale-funnel as repository."

I'm going to give the -funnel fork a chance instead of the non-funnel fork but you can remove -funnel if you just want the regular version

eternal inlet
#

did you get anywhere with this @rose wing ? I'm seeing the same problem

rose wing
eternal inlet
#

Can you explain more? Do I need another integration?

rose wing
#

It just added itself iirc.

#

If you've connected it, check if you now have a Tasmota integration.

eternal inlet
#

I have a tasmota integration, but it doesn't show any entities

rose wing
#

That's what mine did, they connected and went into that integration rather than MQTT like my older plugs.

eternal inlet
#

odd, mine are connecting as I can see it in the console on each switch, but nothing is showing up in the integration or entitiy

#

mqtt.dump also returns a good set of data when I test it

#

it feels like the broker configuration is wrong, it isn't scanning for the topic or something

#

if i listen on the mqtt integration, I see the messages being published by the switches, so they are hitting the broker, feels like the mqtt integration isn't discovering properly...

#

ok, one has turned up now, but it's not working... will leave for a bit and see if some auto magic happens...

chilly olive
#

how to add second device to mqtt?

supple grove
#

Elaborate

chilly olive
#

How to add mqtt server? I have mqtt server details, how do I add it?

quasi scarab
#

I'm having an issue with Studio Code Server using excessive CPU resources. I've been noticing that my Hardware CPU has been pegging out lately. I've been working with Frigate, Deep Stack, and Double Take and have been monitoring my CPU usage. Today, I updated HA,

#

Home Assistant 2023.3.6
Supervisor 2023.03.2
Operating System 9.5
Frontend 20230309.1 - latest

#

and now, my CPU is regularly pegging. I traced the problem to Studio Code Server, which is utilizing 47%-73% CPU.

supple grove
#

Might make sense to investigate the processes in the container with docker stats (use the SSH addon for it) or via the glances addon.
Does restarting the addon fix it?

supple grove
carmine raven
#

G'day! Any suggestions for a reliable and easy to set up timeseries database? I was tempted by InfluxDB, but I am a bit concerned that it is abandoned since 2021...

chilly olive
sly kindle
#

And it looks like supla is too. You only need one broker

young maple
#

Anyone here using Mariadb with HA notice a recent change that would have impacted direct queries that would have worked last month and no longer do?

steady niche
#

I'm trying to set up Frigate and running into the "green image" issue. Where do I check the logs as I see I should be doing?

#

I have reolink 820a cameras and I have tried to set my config up using https

quasi scarab
supple grove
#

The instructions are made with that in mind. You can also run top -c -d1.5 in the terminal of Code Server which is probably even more useful.
Glances is very generic but might still help. You really want something specific for this though.

lunar ore
tawny breach
#

I asked this in the frontend channel but thought since I am asking about an add-on, it may be better fitted here.
Is there anyways to change a color light back to standard white, using the mushroom card? If not, how are people accomplishing this? Seems like being able to go back to a standard white light is a valuable feature. I tried playing with the RGBs but couldn't find anything that resembles the standard white light.

supple grove
tawny breach
compact terrace
quasi scarab
chilly olive
void urchin
#

Does anyone know an addon for document Management? I want to have my incoming Letters, invoices,.. OCR Scanned and extracted and at its Best with matched Company or Person... Does anyone have such an addon or know a good Tool?

I have found docspell but Did Not try it out now...

supple grove
#

Sounds like paperless

void urchin
#

Yeah, some Tool like this, do you have it Up and running?

supple grove
#

No I don't use it

versed marsh
#

I'm trying to install the Nginx Proxy Manager on my HA instance which is running the appliance but I'm getting this error message when I try to install it:

Failed to install add-on
Can't install ghcr.io/hassio-addons/nginxproxymanager/amd64:0.12.3: 404 Client Error for http+docker://localhost/v1.41/images/ghcr.io/hassio-addons/nginxproxymanager/amd64:0.12.3/json: Not Found ("no such image: ghcr.io/hassio-addons/nginxproxymanager/amd64:0.12.3: No such image: ghcr.io/hassio-addons/nginxproxymanager/amd64:0.12.3")

Rings a bell to anyone?

#

Not sure why il's looking up localhost

vital bear
#

DuckDNS question: I've got it setup to see my duckdns on local, but it breaks when trying to do remote. I'm fairly confident it has to do with my port forwarding, which I tried to setup as (see below). Am I missing something obvious? My networking class for college isn't until a few weeks haha

vital bear
supple grove
#

Upload to a image site and post a link to it here instead

vital bear
#

Will take a look, thanks!

fervent harbor
#

hey all switched server to something 12gen based and I can not get frigate to use igpu on my 12th gen any ideas (I do see in hardware info no vga is detected now and just have DEVNAME: /dev/vga_arbiter DEVPATH: /devices/virtual/misc/vga_arbiter MAJOR: '10' MINOR: '127' SUBSYSTEM: misc it might have been this before, just never needed to check before on older celeron based setup as it worked using hwaccel).

fervent harbor
#

2023-03-25 09:03:02.507804399 [2023-03-25 09:03:02] frigate.util ERROR : Unable to poll intel GPU stats: No device filter specified and no discrete/integrated i915 devices found

weary stratus
#

I try to add influxDB and run it. It installs , when I start it, I see things happening in the logs, but then nothing more happens (last log message: ... INFO: Starting NGINX ...). So the webserver is trying to start, in order to enable me to push 'open web ui', I guess, but it stalls. Can any of you do me a favor, just try to install and start influxDB addon to see whether it works for you. I need to know whether it is just my installation or an overall problem.

#

Any advice is welcome!

supple grove
#

Works fine. INFO: Starting NGINX... is the last message. What happens if you visit InfluxDB from the sidebar?

weary stratus
#

Thanks for coming back, it says: "Add-on is not running. Please start it first"

#

Strange ...

supple grove
#

Strange that the addon's logs don't say anything

weary stratus
#

So, you get an interactive web page, where you can select "open web ui"?

#

Anyone else, please?

supple grove
#

I add it to the sidebar and click there

weary stratus
#

and it opens?

supple grove
#

Works fine

weary stratus
#

Pffffffffffffffff

supple grove
#

If you check the InfluxDB addon's overview where the Uninstall button is. Does it say it's running? It will show CPU and RAM usage if it is and there should be a Stop instead of Start button. When an addon is stopped the log should also be empty, usually

weary stratus
#

Let me check

#

No, it just shows the original screen with "start" on the left, "uninstall" on the right, and on the top right a red dot (indicating add-on is stopped), Pushing START AGAIN, doesn't do anything, logs also don't change any more. All I can do is uninstall, install again, start it, see the logs move on until shows INFO: Starting NGINX again. (endless loop!)

supple grove
#

Not sure

weary stratus
#

Wait, same thing happening for other Add-on's as well, so it is not InfluxDB related ... will restart everything and check again!!!

#

Problem seems solved! @supple grove Thank you very much for taking the time to react!!!

supple grove
#

No problem

jaunty elbow
#

Hello 👋 Has anyone faced issue with the core MQTT addon not starting due to the following error: 2023-03-25 13:51:42: Opening ipv4 listen socket on port 1883.
2023-03-25 13:51:42: Opening ipv6 listen socket on port 1883.
2023-03-25 13:51:42: Opening websockets listen socket on port 1884.
2023-03-25 13:51:42: Error: Unable to create websockets listener on port 1884.
[11:51:42] WARNING: Halt add-on
s6-rc: info: service legacy-services: stopping

#

There's nothing running on port 1884 and until last night everything was running fine. I have restored a backup, but no luck. MQTT just won't start 🤷‍♂️

supple grove
#

How did you check that there's nothing running on 1883-1884?

jaunty elbow
#

sshd into the VM and ran netstat

#

➜ ~ netstat -tanp|grep 1883
➜ ~ netstat -tanp|grep 1884

#

Returns nothing

#

VM is running bridged with it's own IP, so host machine ports should not matter

#

Have already tried uninstalling the addon, no change, reverting to default settings (which they were already), no luck.

#

Tried changing ports from the addon configuration, but that didn't work either, it still defaults to standard (apparently unusable) ports

#

Not sure if this is caused due to the supervisor update to 10.0rc1

split stratus
jaunty elbow
#

Seems to be common issue, downgrading to 9.5 seems to be solution

split stratus
jaunty elbow
#

Most likely it is ha core/supervisor issue

#

But I am downgrading the OS now as someone had luck with that

split stratus
jaunty elbow
#

ha os upgrade --version 9.5

#

It's taking uncomfortably long time to reboot 😂

split stratus
#

I am reverting too

#

after entering the command

# 
# ha os upgrade --version 9.5
Command completed successfully.
#

It is rebooting now 🙂 and up and MQTT is up and running again!

jaunty elbow
#

Yep, can report the same

#

It seems now my Z2m with Skyconnect is borked 😑 what a wonderful day with HA 😅

supple grove
#

Snapshots/backups are very useful with VMs

jaunty elbow
#

I do have daily backup on HA, but not on the VM itself

#

Will do that once I get my NAS with proxmox

supple grove
#

Wasn't meant as a jab. It's just one of the cool things VMs allow you to do easily and provide a huge benefit

jaunty elbow
#

Oh wow, what a journey 😂 So the OS 10.0rc1 borked the MQTT, which broke Z2M. Before figuring out that MQTT was broken, tried few things to get Z2M with Skyconnect to run. Tried the SL Multiprotocol addon which apparently automatically flashed the Skyconnect FW. This FW doesn't work with Z2M ezsp as it seems, so needed to flash new Zigbee only FW to Skyconnect. Finally Z2M and MQTT works again and my home is functional 😂

jaunty elbow
jaunty elbow
green dune
#

Anyone familiar TaailScale VPN and Mosquitto MQTT ? I'm trying to get Mosquitto MQTT to connect over Tailscale network. I think I need to add a route but I have not seen any reference to doing that yet. The Tailscale interface and ip don’t seem to be available in the HA user space so I’m not sure how to reference it.

chilly olive
#

Can someone help me with mqtt? I've been trying to do this for a few days now... I currently have only hass agent connected via mqtt to send information and it works, but I don't know how to add another supla device that has its own broker server.

uncut flume
#

My Mosquitto broker won't start anymore since the latest OS version that was released today. The error that I get is: Error: Unable to create websockets listener on port 1884.

Any idea how to fix this?

brittle night
#

something else trying to use that port?

uncut flume
#

I didn't install anything... I only updated the OS version today. before that everything ran fine. Any idea where I can check what apps use that port?

#

Just heard that there are issues with the new OS version and Zigbee2MQTT. I have to downgrade to 9.5

lunar ore
#

@raven tinsel thank you a lot for updating tailscale, you rock!

raven tinsel
#

Today I've mostly put in place automated dependency upgrades for almost all add-ons (which is why it bumped on Tailscale)

violet tapir
supple grove
harsh mortar
#

Anyone know how to edit files for addons? I'd like to make some changes to the data directory specifically adding a custom_js.html file

heady lava
#

any known issues with the add-on screen just being blank when clicking into that part of Home Assistant?

viscid orchid
#

Clear your browser cache

heady lava
#

still blank

#

Uncaught (in promise) SyntaxError: "" literal not terminated before end of script entrypoint.fc042133.js:1214:46481

#

@viscid orchid

kindred escarp
harsh mortar
#

I have, with the addons shipped with hassos they're not available in /addons to be edited. @kindred escarp

heady lava
#

basically, the addon section won't load and no addons will. all giving that same error in Dev Tools.

violet tapir
#

I installed the samba add-on now. but it doesn't work - no configuration available. and also it's not listed in the installed plugins. but it's there and I can click on start, however then of course it complains about missing username and password, which I cannot set due to the missing configuration. Can I edit a config file somewhere?

violet tapir
wild basin
#

Hello !!

#

Where can I find the debug file of a hassio addon?

zinc terrace
#

Frenck just discovered Renovate? Haha

zinc terrace
#

There's a log tab on all add on pages. Otherwise maybe you can provide a little more info on what you're looking to do specifically or what your problem is

zinc terrace
wild basin
#

I found in the end I had misconfigured the logs ^^

raven tinsel
#

So, was very familiar with Renovate, including using it in enterprises

zinc terrace
#

Last I remember people were moving to renovate for less PRs. I assume github would he working hard to keep dependabot competitive though

#

I feel like there was change recently specifically to reduce them for dependabot

raven tinsel
#

I like to use native options if available, hence I switched to dependabot back in the days

zinc terrace
#

Yeah absolutely

raven tinsel
#

Well, dependabot isn't actively gaining new features

zinc terrace
#

That's surprising

raven tinsel
#

And being able to write your own update logic in Renovate

#

works very well for my use case now

#

If it becomes available in Dependabot, I would not hesitate to switch back again

brittle night
#

Does this mean we should see more frequent updates for addons?

zinc terrace
raven tinsel
#

No, this has nothing to do with releasing.

zinc terrace
brittle night
#

I figured if it made less work for you then maybe you’d be able to use that recouped time for improving them / other projects as well

raven tinsel
#

yup

zinc terrace
#

Have you written any custom updates Frenck?

raven tinsel
#

yes, tons

#

basically it is all I use

zinc terrace
#

Any examples you could point me to? Or is it it just in the config files. Sorry, would research more myself but on mobile atm

raven tinsel
#

One the simpler examples, it updates external repos/dependencies, but also updates version pins of packages in Dockerfiles

#

and handles base images in the add-ons configuration as well

zinc terrace
#

That's pretty cool. I gotta add this to my projects. Always thought it would be nice to update versions where I've like parameterized it in a curl call

raven tinsel
#

Oh, and something dependabot doesn't do, but Renovate does I noticed already...

#

updating in the weekend 😄

#

For some reason, dependabot doesn't do normal updates during the weekend

#

So on Monday morning, when I wake up, my inbox is usually exploded with bumps

zinc terrace
#

That's cool, so you can specify sources as well. I assume that would be like github releases, npm, etc

brittle night
#

Feature not a bug smile

raven tinsel
#

It is annoying, as my monday morning the first 2 hours is wasted on processing notifications and merging dependencies

#

Which is just an awful and demotivating start of the week

zinc terrace
#

That's so weird. I would think you could specify when it did the checks

brittle night
#

Sounds like we need to crowdfund you a peon

raven tinsel
#

LOL I can also just unsub and leave those dep bumps to rot until I get to them when I move by

#

But my brain isn't capable of that

zinc terrace
#

Do you have any projects set to auto merge on a successful build?

raven tinsel
#

yes

brittle night
#

Isn’t that what all the betas do?

zinc terrace
#

That would make sense

brittle night
#

While we’re posting memes in #beta waiting

raven tinsel
#

e.g., if I bump my base image now, with a new patch version

#

all add-ons will get a PR, and auto merged when the CI passes

#

which just saves me creating 40+ PRs manually

zinc terrace
#

Trustnoonenotevenyourself.jpeg

#

That's cool though. Must be satisfying

raven tinsel
#

rofl, it doesn't release. I still need to do that manually (I always test each release manually as well)

raven tinsel
zinc terrace
#

Yeah. Too bad I'm in a train cuz I'm motivated to set this up now lol

raven tinsel
#

I want to create re-usable rulesets at some point, but this, for starters, is good enough

zinc terrace
#

I was doing that manually for a lot of my personal projects for a while but turned it off cuz that was too annoying. But it was crazy how often stuff broke stating at the latest version. Would have been so annoying to have to do that all at once

zinc terrace
#

Really liked your HA config checker action btw. Used it a lot to write my own custom one

raven tinsel
#

Rofl, that thing became so simple/small eventually

#

right now it is a composite action running a container 🤷‍♂️

zinc terrace
#

Only hacky thing I've since come to terms with is I have to manually clone my custom components so it doesn't fail

raven tinsel
#

oh right, that is still a thing 🙂

zinc terrace
#

Not sure there's really any way around that, but it's been a while

#

But I really should focus on getting my integration's tests finished so I can submit the PR. I couldn't get the tests running in intellij and it was bugging me. Still can't figure it out. Have spent a few sessions on it 😑 Everything is finished except that, including brand and docs.

raven tinsel
#

🤷‍♂️ Just run pytest from the command line

zinc terrace
#

Something about that tests/typing.py file is screwing things up

raven tinsel
#

(at least, that is what I do; don't use any IDE for stuff like that)

zinc terrace
#

Yeah, just wanted debug lol

#

I don't usually either but especially while I'm learning things, debug would be helpful

#

Haven't written python tests in years

#

Only reason I could attach a debugger to the service while writing the integration was by deleting tests/typing.py lol

#

Intellij is getting confused because it conflicts with the built in module.

brittle night
#

i just noticed when doing the tailscale addon update there was no "what's changed" with any details just blank

zinc terrace
#

But anyways, sorry, well off topic

brittle night
#

other addons all show the changes

zinc terrace
#

Yeah some add-ons just don't do it

#

I'm guessing it's some feature of the update integration

raven tinsel
brittle night
#

maybe a browser related issue?

raven tinsel
#

🤷‍♂️ could be

brittle night
#

not worth diagnosing just brought it up in case there was an issue on your end

raven tinsel
#

hehe the changelog system has been working for years now 😄 I'm pretty confident in that one 😄

#

(although I have to update Python for it, which it always scary :P)

#

I touch it the least possible 😛

weary stratus
#

Hi all, long time user but recently installed AdGuard add-on. Works great but have one issue. I’m using UniFi network gear and has my dhcp server as well. I am pointing dns to my static home assistant IP. It’s working however all of the “clients” on AdGuard show up as 192.168.0.0 — unless I set the dns server manually on a device to point to my home assistant AdGuard IP. I’m sure it’s something stupid I’ve overlooked. Any thoughts? Thanks

#

On the list of runtime clients, they’re all listed, but doesn’t seem to be resolving for the actual client list

brittle night
weary stratus
#

Thats what I have.. AdGuard then 1.1.1.1 .. clients are connecting properly through AdGuard filtering, but AdGuard itself doesn’t break our individual client ips. All going through 192.168.0.0

final prism
#

Anyone update to 3.2.3 Tautulli? Wont start and crashes for me 😦

#

[10:51:34] INFO: Updating running configuration...
/etc/s6-overlay/s6-rc.d/init-tautulli/run: line 37: TAUTULLI_VERSION: unbound variable
s6-rc: warning: unable to start service init-tautulli: command exited 1
/run/s6/basedir/scripts/rc.init: warning: s6-rc failed to properly bring all the services up! Check your logs (in /run/uncaught-logs/current if you have in-container logging) for more information.
/run/s6/basedir/scripts/rc.init: fatal: stopping the container.

raven tinsel
#

That sounds like a bug that I might have introduced

#

going to look at it right away

brittle night
#

Frenck introducing bugs?! Time to go buy a lottery ticket

raven tinsel
#

lol I do

#

I'm still just human

#

Oh right facepalm

#

I see the issue, will fix it @final prism

brittle night
#

Legen… wait for it… dary

weary stratus
#

I figured it out. Apologies. My router has WAN and LAN settings. The latter of which was sort of hidden. Works great now sorry

brittle night
final prism
brittle night
#

@raven tinsel is there somewhere we can request specific addons from you?

raven tinsel
#

I do have some I want myself still (and also haven't created yet)

brittle night
#

am i special enough to make an exception for? smile

#

haha nah i understand

raven tinsel
#

hehe no, I have tons off stuff I still need to do

#

adding to that, will not solve anything

brittle night
#

frenck cloning machine inbound

#

yeah makes sense

#

it is just nice to have the high security rating for addons

#

as far as trust goes

#

makes me feel all warm inside

raven tinsel
#

Nevertheless, curious 😄

#

what would you like to have?

brittle night
#

hyperion / hyperHDR

#

i've used both, currently using hyperHDR, and they both run great as addons instead of needing to run hyperbian or in a container on my server

#

i have a d1 mini with wled behind my tv and using android hyperion grabber to send data to hyperHDR to control the d1 mini

#

really fun setup

raven tinsel
brittle night
#

i figured since you are a wled fanboi you might be a bit more inclined

raven tinsel
#

hehe no, sorry

brittle night
#

no prob

raven tinsel
#

These things are cool, but also very fragile

brittle night
#

howso?

#

you mean the addon itself or the whole thing requiring a d1 mini etc?

#

it has been set and forget so far (luckily)

raven tinsel
raven tinsel
#

Quindor sent them to me, super helpful for testing on my desk

brittle night
#

sweet

raven tinsel
#

with various setups

brittle night
#

holds back asking for a wled integration update featuring cct support

raven tinsel
#

progress is still made 🙂

#

couple of PRs merged again for it this month

brittle night
#

i havent looked since the last time we chatted

#

didnt want to bug you since you said it needs tests done that i cant help with 😦

#

it's a bummer we're going on 13 months since it was removed but i understand you gotta focus on other things

raven tinsel
#

Things have review/cycle time too

#

my last PR for rewriting platform tests took 2 weeks (for a single platform) before it was approved and merged 🤷‍♂️

brittle night
#

i asked aircoookie if he could help test like 6 months ago and he said he'd look into it or talk to you or something

#

i dont really hang out in wled discord anymore but last i checked he wasnt really active anymore

#

just blazoncek which has an attitude lol

#

anyways i still appreciate all the things you do. crazy how often your name and a few others comes up with hass stuff

raven tinsel
#

🤷‍♂️ Not that hard, I work on HA fulltime

silk root
#

how to use the remote SSH VScode?

raven tinsel
#

@silk root Did you read the documentation?

#

🙂

silk root
#

Visual Studio Code Remote - SSH
Setting the following parameters as is shown will allow you to connect to your Home Assistant instance using VSCode Remote - SSH:

feral hornet
#

hello, I am under the latest version of HA and the plugin. All my devices are well recognized.

however for floodlights unifi Protect the control entity is not available, only sensors and diagnostics are used.

thanks

brittle night
#

is it possible to skip esphome addon updates like you can for other updates?

#

i'm not seeing an option to in the gui

sly kindle
#

You’re not seeing the skip option when you check the update?

brittle night
#

No

sly kindle
brittle night
#

Only install here

sly kindle
#

Hmm. Cache issue?

brittle night
#

I’ll try reloading browser

#

i dont want to clear cache 😦

#

doesnt show up on edge or firefox (without clearing cache)

#

theme is set to backend-selected to rule that out

lime crescent
#

(i'll just paste it here: hello guys. I probably just need the pointer to a good tutorial or the right docs but I am really lost on this one: I want to switch my MQTT communication to TLS because I will have a few 'road-warrior' devices that will report to my HA instance from outside my home network. I am running a DynDNS, I have all the knowledge to port forward 8883 to my HA instance, etc.
I even would call myself a pretty technically inclined person, but I cannot for the life of me figure out how to setup HA with core-mosquitto to switch to port 8883 and use TLS for the connection. For my purposes I would not even need cert verification on both ends, PW auth is fine for me, but I want the encrypted channel. If certs can be done easily that is fine too. But I can't figure it out. Also the docs confuse me where to configure stuff, I cannot see any config entries in configuration.yaml etc.
Can someone enlighten me please?^^
[15:36]
I should add: I'm running the HA supervisor VM on proxmox as delivered by HA and have mosquitto running in non-TLS mode within my network for months to monitor and control some power plugs. Everything is up-to-date and latest greatest.)

#

As far as I could figure it out there is 2 aspects: The mosquitto config itself and the config to connect mosquitto to HA. The latter can be unencrypted as it is literally within the same VM where HA is running.
So I guess what I need to do is tell Mosquitto to accept TLS connections. I tested my device using 1883 and this works fine, but what exactly do I need to do to get to a full MQTTS setup? In my device I can give a root CA, I can tick to enable TLS and set the port to 8883. But what exactly are the settings required for mosquitto?

silent sage
#

HI, i am new with AdGuard, and my new installed aduard only listen to 127.0.0.1 (setup on HAOS)

silent sage
errant onyx
#

@lime crescent is it possible to throw together a VPN instead?

#

would save you some thinking about how to expose the services securely

silk root
errant onyx
#

maybe not the right channel, but did you get the HA vscode extension to work locally?

lime crescent
#

after I figured out that the setup interface in the mosquitto integration actually sets up comms bewteen HA and Mosquitto and has NOTHING to do with the comms between Mosquitto and the actual clients, I was able to follow this external guide http://www.steves-internet-guide.com/mosquitto-tls/ and create the certs, make the settings in the mosquitto add-on GUI and then it immidiately started working

#

I can actually not believe something so simple is not to be found anywhere for an add-on almost half of all HA installs use^^ But maybe this is just my personal stupidity xD

#

now I have auth with user+pw over TLS working and just securely forward the port externally directly to Mosquitto and all is well.

formal spindle
#

Hi guys, im trying to add service to existing hacs integration. Basically im getting sensor/switch values from rest api ( every 1min). When I change switch status it trigger post request, which change status of remotly device. But the HA entity will change after full cycle (1min). I know that refresh service can help I found this https://community.home-assistant.io/t/manually-refresh-rest-sensor/353208. But how can I implement it into plugin? Is there any developer's documentation for existing hacs/hass methods where can get more knowledge? Or do you know any simple hacs integration where they solve this issue? I will be gladfully for any type of help 🙂

hidden cove
#

Hey everyone. I'm trying to migrate data from SqLite to MariaDB but I'm not able to connect to mysql/mariadb from the command line. The addon is running and I can't see any issues from the Log. My installation is a VM installation. I'm probably just too new to HA to get behind the whole supervisor thing. It looks like there's no real mariadb process running in the system so I'm not quite sure on how to connect to mariadb. Any ideas on what I'm doing wrong?

errant onyx
#

can you connect using phpmyadmin?

hidden cove
#

never tried - I guess i gotta install that first as an addon, right?

errant onyx
#

might be worth just to see if you can connect to it. I mean the only config needed for mariadb was like username/pw/access

hidden cove
#

nevermind - got it working now. I accidentally used the wrong parameter when switching around hostnames. core-mariadb is just working fine.

#

thanks for helping anyways 🙂

untold oak
#

hey, building an addon, but can anyone tell me how to make an optional schema?

#

it always complaints about the subsection

#

this doesnt work

# 
schema:
  doorbells:
    - name: str
  mqtt:
    host: str?
    port: int?
    ssl: bool?
    username: str?
    password: str?        
  system:
    log_level: match(^ERROR|WARNING|INFO|DEBUG$)
    sdk_log_level: match(^NONE|ERROR|INFO|DEBUG$)?
#

it compiaits about the missinq mqtt: entry in options, if i define it in options , the n it work, but then its not optional anymore

#

Failed to save add-on configuration, Missing option 'mqtt' in root in ...

brittle night
untold oak
#

ok

steady echo
#

has anyone noticed high memory usage on frigate add-on. Frigate memory usage will start climbing to over 6GB and only fixes itself if I restart the add on

pallid walrus
#

I have 4 ESP-32 devices flashed with ESPHome. All 4 of them show up in the ESPHome Add-on, but only three of them show up in the ESPHome integration. I've reinstalled it several times but it still does not show up. What am I missing?

raven tinsel
hidden cove
raven tinsel
#

Yes, because MariaDB is the better database on a longer term.
👆 I wonder what you mean by that? Got an example?

#

what is the concern in Home Assistant in this case?

#

As we are actually strongly recommending SQLite

#

Not sure what you mean by „not optimized and performant for HA“
That MariaDB will perform worse for Home Assistant. It will make your setup slower

#

Home Assistant is optimized for using SQLite

errant onyx
#

MariaDB with the recorder config is just about the history data I presume?

raven tinsel
#

state/event history and long term stats, yes

errant onyx
#

I guess it'll really depend on how much data is flowing but for a small use it's probably going to be just the default is best

#

You probably don't want to be loading up an entire dashboard of charts off it but then again you should be offloading the data elsewhere for that anyhow

raven tinsel
#

You can store data for years without issues

hidden cove
# raven tinsel > Yes, because MariaDB is the better database on a longer term. 👆 I wonder wha...

I need to underline that I'm not trying to tell anyone to move to a relational DB at all. That's just a personal decision and I don't want to say that anything else is bad.

I got a bit of history working with databases and I just prefer real database engines over sqlite for stuff that runs longer and grows over time.

I totally get why SQLite is the default and I'm sure there are tons of situation and reasons where it's better to stay on SQLite (e.g. if you're running on a Pi from an SD-Card which I'm not). But other than that I can't see why Homeassistant should be "not optimized" for SQL but optimized for SQLIte. Can you explain that?

raven tinsel
#

Hehe yeah, I'm an certified MySQL administrator and developer, I'm well aware of MariaDB and its inner workings 🙂

#

it was my job for years

#

I will not run HA on MySQL 🙂

#

Can you explain that?
HA is specifically developed against SQLite. Everything done to optimize performance, checks and everything else. It aimed towards SQLite

errant onyx
#

I was gonna say, it's what it's bundled with so they'll optimize against it

raven tinsel
#

Not just that, backups are bundles with the version "snapshot" of core to match (dbs are not backwards compatible when restoring backups)

#

and tons of more of those little reasonings

#

but, that said MySQL is supported; but rarely a better choice

#

(even when running on an SD card)

#

Anyways, gotta run, appointment outdoors 👋

hidden cove
#

especially not a good idea when running on SD Card 😉

raven tinsel
#

Doesn't matter much anymore

#

As said, HA optimized for the use of SQLite (including SD cards)

#

lots has changed over the years

hidden cove
#

running a rel db engine on SD cards is nothing I would suggest to anyone.

#

that's a sure SD killer sooner or later (maybe sooner)

raven tinsel
#

SD cards will die eventually... only way to win this game, is not to play.

#

Still, even medium sized HA instances on a Pi4, should last years now

errant onyx
#

SD cards are definitely a losing proposition

#

I wonder if they would start shipping with eMMC but that's a topic for another day/channel

raven tinsel
#

EMMC, also has wear 😉

#

So do SSD//NVMe drives

idle escarp
#

Are there any known problems with the api intergration within HA? I have two ttgo t-display (Esp32) devices and I flashed both with multiple codes fro ephome. Everything works, WIFI, sensor, web server, display (without no difference, withou sensor attached no difference) but it cannot do discovery. Even a manual one time out. Settings it to manual IP makes no difference. Rebooted AP and HA (Inc. VM) Changed computers, flashed esphome through webinterface, through esp_flasher, removed the integration, re-added it with different name, I am gettin crazy sinceIjust want the values, two other devices are discoverable if that means something

vast grove
#

I am using the influxdb add-on, which creates an HA database. I pointed to an existing influxdb instance I already run on the network. I want to also utilize a different database within that same influxdb instance in order to create "sensor" data. Is this possible?

fervent raptor
#

Hello. I need some help. I can't resolve my issue with installation add-ons on HA. That what it paste: "Can't install homeassistant/amd64-addon-configurator:5.5.0: 500 Server Error for http+docker://localhost/v1.42/images/create?tag=5.5.0&fromImage=homeassistant%2Famd64-addon-configurator&platform=linux%2Famd64: Internal Server Error ("Get "https://registry-1.docker.io/v2/": net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)")"

slim ferry
#

400: Bad Request when I want to access HA via Cloudflare

tough delta
#

is the ssl certificate the cloudflared issues downloaded to HA? if so, what is the path? i am connecting locally in some cases using internal dns and not public cloudflared one and believe I need to point my ssl cert paths appropriately. its not in the typical ssl directory

gray depot
#

Hey, this may be outside of a lot of peoples wheelhouses too but at this point I am desperate.

So basically my homenetwork is behind CGNAT, so I use cloudflared to have a tunnel and to be able to access my home assistant from remote. Now I'd like to also have a vpn running at home so I can ssh into my home computer when at work. I only have one pi and saw that there is a wireguard addon. So I tried to somehow have the wireguard addon also use the cloudflared tunnel to access the internet. But I am pretty much stumped at how to actually do that. Does anybody have any suggestions for what to try or perhaps even experience?

river coral
errant onyx
brittle night
errant onyx
#

I heard cloudflare has some kind of zero trust thing but not sure how that works without trying it

brittle night
gray depot
gray depot
#

sorry, the first reply was supposed to be for blue's message

brittle night
#

./giphy how-dare-you-greta.jif

gray depot
brittle night
gray depot
#

hm, yeah setting up a separate vpn client doesn't really work for me bc cgnat and lack of more computers

solid wedge
#

Is there a way to build an addon with multiple ingress ports?.... I want to use home assistant to access a bunch of my internal services. I was going to iframe them, but they aren't accessible from outside via URL. So I was thinking about making an addon proxy using ingress. But ingress only allows a single port ... Is there another way to do this? Basically I just want an icon for the services I'm not running as addon's....

#

Alternative is just to make a custom addon per app I want to proxy

normal lagoon
#

Have a question about the mqtt server addon configuration. I have configured server and client certificates and they seem to work fine, but I am still required to enter username and password.

Configuration:

logins: []
require_certificate: true
certfile: server.crt
keyfile: server.key
customize:
  active: true
  folder: mosquitto
cafile: ca.crt
use_identity_as_username: true
debug: true
bitter skiff
placid galleon
#

Has anybody else had the zigbee2mqtt addon seemingly die during an HA restart?

#

More than once now I've restarted HA (not the entire host) and all my z2m entities report unavailable until I've restarted the addon

brittle night
#

My z2m addon always stays up

lean fox
#

no problems with z2m here.

placid galleon
#

Bizarre, it's only started happening recently too, my stuff has been stable for ages

#

Tho I did change to EMQX recently but I don't see how that could cause this

bitter skiff
#

Months ago my Z2M used to do that, it was fixed though. Have you checked out the logs?

placid galleon
#

I looked at the addon logs before I restarted and there wasn't any errors going on

#

I just looked at the state history and there is a gap for the last reboot followed by showing as unavailable immediately.

#

Is there some retain MQTT setting I could be missing?

bitter skiff
#

Was there anything in the add-on logs before you restarted Z2M?

placid galleon
#

Just the normal state change chatter that I saw

bitter skiff
#

In the add-on logs though

placid galleon
#

Here's 3 random devices after a restart

#

all zigbee

#

So it's like not showing sensors that haven't sent an update

#

But mains-powered stuff is updating right after the reboot

bitter skiff
#

Ah, so Z2M is running

placid galleon
#

Yeah, it's seemingly just my battery-powered zigbee devices reporting unknown after a reboot

#

Which is bad, my furnace is driven off an average of multiple sensors but half of them are zigbee battery powered ones so my average gets skewed after a reboot

bitter skiff
#

Could see if #zigbee-archived has any thoughts... odd behavior though, my issue earlier was Z2M wasn't starting at all and had an error.

placid galleon
#

Yah since it's not the whole addon it seems like some z2m quirk somehow

bitter skiff
#

Suppose there could be differences between brokers as well, as you said you switched, also seems odd though

placid galleon
brittle night
#

Is there a different last will default on that broker over mosquito?

placid galleon
#

Good question lol

brittle night
placid galleon
#

I honestly know very little about EMQX still haha

brittle night
#

Idk shit about it just a thought since last will and testament is built into brokers

placid galleon
#

I suspect it could just be a matter of how quickly it deletes data when the last subscriber disconnects

#

That HA could restart within the default mosquitto window but not the EMQX window

brittle night
#

Tag me with whatever you figure out I’m curious