Plenty of errors, but HA loads and works. What ca I do?

Configuration
1

Hi

I’m on an Intel Nuc i3 with Proxmox and HA installed in docker on Ubuntu Server.

## Home Assistant 0.109.3
Versione Frontend: 20200427.2 - latest
Interfacce Utente personalizzate: [CustomUI-fix-nav](https://github.com/rccoleman/home-assistant-custom-ui): 20190322
## Supervisor Version 220

From a couple of days, I’m getting a lot of issues and my log is full of errors, but HA loads and I can see the dashboard.

I don’t understand where I can check for errors or what is causing them

Thanks

Here the log.

2

Start with fixing the problem with your database, then paste any further errors.

1 Like
3

Thanks @anon43302295

I’ve removed now maria_db

recorder:
  purge_keep_days: 7
#  purge_interval: 1
#  db_url: !secret mariadb_url

Here is the new log, but still a lot of errors

4

Actually, most are warnings. Home Assistant always posts a warning message when it loads a custom integration and your system has over a dozen of them.

The error messages concern the integrations it simply could not find.

2020-05-05 14:31:56 ERROR (MainThread) [homeassistant.setup] Setup failed for meross_cloud: Integration not found.

as well as integrations that failed to connect:

2020-05-05 14:32:02 ERROR (SyncWorker_8) [homeassistant.components.broadlink.switch] Failed to connect to device

2020-05-05 14:32:03 ERROR (SyncWorker_6) [homeassistant.components.orvibo.switch] S20 at 192.168.1.98 couldn’t be initialized

and integrations you didn’t configure correctly:

2020-05-05 14:32:14 ERROR (MainThread) [homeassistant.components.sensor] The glances platform for the sensor integration does not support platform setup. Please remove it from your config.

and integrations that you’ve duplicated:

2020-05-05 14:32:14 ERROR (MainThread) [homeassistant.components.sensor] Entity id already exists - ignoring: sensor.date_time

and integrations that have simply failed to work:

2020-05-05 14:32:14 ERROR (MainThread) [homeassistant.components.media_player] Error while setting up enigma platform for media_player

Start fixing them, one by one.

1 Like
5

Indeed, seems a lot of network failures as well.

1 Like
6

@123

Meross integration removed from HACS Integrations, but not able to remove from Settings->Integrations
Broadlink removed from configuration, sensors and switches
Orvibo removed the IP not available
Glances removed
sensor.date_time probably solved
enigma media_player changed IP

Still some errors, but less than before
Here

I don’t understand the ones with “Platform automation does not generate unique IDs” because if I search all over /config the entities ID in the errors, I don’t find them.

What do you mean by network failures?
What can I check?

Thanks both!

7

Many of the errors (I haven’t had chance to look at your most recent log yet) we’re ‘X cannot be reached’ or ‘error contacting X’ type errors.

This would indicate that there is a problem with the network, maybe Internet connection or internal routing/switching. Without knowing your network set up I don’t know where you would start checking but I’d certainly be rebooting my router and making sure I can ping all the devices I was expecting homeassistant to talk to.

1 Like
8

If you have more than one automation with the same id then that’s the message Home Assistant will report. It attempts to load the duplicate but fails because it cannot give the duplicate a different id (" … does not generate unique IDs").

The same situation is true for duplicated entities and explains why you’re seeing errors messages for sensors and switches.

It’s like you have entities and automations already registered (in the .storage folder) and then you’re trying to load second copies of them from YAML files.

9

Thanks

Deleting .storage could help?

10

If you stop homeassistant, delete your database file and your .storage directory but keep your configuration.yaml and associated yaml files, it should effectively ‘factory reset’ your install, but with your configuration still in tact.

Please double check this though as it is a bit of a nuclear option and I’m not home right now to double check the files!

1 Like
11

Last thing.
I’m not able to remove Meross Cloud from Integrations (Settings->Integrations)
When I click DELETE a popup shows up with “Are you sure?” but after clicking Yes nothings happen.

Is there any manual way to remove it?

Thanks

12

UI Integrations are kept in storage, so if you go the nuclear option this is another thing that will be swept away. Of course you’ll have to re-add your other UI integrations once you’re back up but it’s trivial.

I can confirm that I was right about the files, so if you’re not precious about a factory reset then that’s what I would do…

Stop homeassistant > delete database file > delete storage directory > reboot

When it comes back up you’ll be at onboarding. Add your user and password to get to the frontend and you will find that all your yaml based configurations are already there (possibly with a few errors if automations have already tried to run that reference entity_ids that are from ui integrations).

Then go in to the integrations page, add back your ui ones, restart one more time and see what the logs say, but by this point you’ll either be error free, the errors will reside exclusively in your yaml and can be corrected easily or the errors are due to an external factor such as network hardware etc.

Please note that if you have a particularly awesome lovelace set up in normal mode, that will also disappear with the storage directory, so you’ll need to back that up (personally I would just copy the yaml from the gui editor in to a text file on my laptop and then copy it back over after the reset).

1 Like
13

Thanks very much for the explanation!

1 Like