When system starts up in recovery mode all integrations are lost for good

[mattinglis]

The problem

When the system starts in recovery mode, it looses all configured integrations. Even once the error is resolved, you must re-add all integrations

What version of Home Assistant Core has the issue?

2024.2.4 onwards

What was the last working version of Home Assistant Core?

unknown

What type of installation are you running?

Home Assistant OS

Integration causing the issue

No response

Link to integration documentation on our website

No response

Diagnostics information

No response

Example YAML snippet

No response

Anything in the logs that might be useful for us?

Set something invalid in the configuration.yaml 

I encountered it on trusted_users being invalid. Which interestingly isnt validated in check configuration.

Restart the system, when it starts all integrations are gone.

Correct the error, restart the system, all integrations are still gone and have to be discovered again.

Additional information

No response

[prat33knz]

Just experienced this myself and stumbled on this recent issue.

[on4ami]

Same here, exact the same scenario

[erasmus83]

Also had this issue yesterday on 2024.3.1 - ended up restoring from backup

[joostlek]

Do you see anything in the logs? Can you check the config/.storage/ folder on your system. I have the idea one of the files got corrupt and then it defaulted to the old one. It usually renames the old file, so it should be there. Please don't directly post it here, it contains a lot of sensitive information.

[njsaunders]

Same thing here - HA absolutely nuked my entire installation after a typo updating trusted network in configuration.yaml. No back. No renamed file. Absolutely gutted - YEARS of configuration lost.

[iT3han]

Got the same issue. After typo in trusted_networks HA started into recovery mode. After correcting typo and restart HA, all integrations are gone. Had to restore from backup.

Core 2024.4.3 Supervisor 2024.04.0 Operating System 12.2 Frontend 20240404.2

Checked the mentioned config/.storage/ location and could not identify a renamed file

[ankurpandeyvns]

Have the same issue. Thanks to the backup I was able to restore

[brjhaverkamp]

Just to add a "me too" I was bitten by this. I just started with HA a few months ago, and lost my first month of data and my initial setup. I wasn't so lucky/smart with backups, so I had to redo the whole setup. ..

[Blackmood667]

Same for me. HA started in Recovery Mode after Power loss, seems there was a typo in the configuration.yaml. Also my Backups don´t restore the Integrations...

Edit: Old Core Backup seems to do the job. Strange.

[posssson]

Same here... and no backup work..

[marloncarvalho]

Same happened here. I can´t still believe that I´ve lost my whole configuration. Tons of integrations, devices, and everything just wiped out. It was just a typo on my configuration.yaml. It happened thousands of times before and nothing like that happened. Please, tell me I'll be able to restore my integrations.

[titanowly]

same, lost everything

[deezid]

same here? WTH

[rapiertg]

Typo in trusted_networks in my case wiped 30+ integrations. Unfortunately this directory were not part of a backup :(

Docker image 2024.6.3 so this nasty thing is still here...

[teji24]

Same Here !!!! :-( with (probably) typo problem in trusted networks/users. restart, loose device, integration, history, .... REM for next users : after some try (& proxmox restore ;-) the main problem is not typo/indent : but use of USERNAME in place of USERID (long number) for trusted user. You have to use USERID (long number), NOT the username. it's in the doc. But not the disaster effect if you make error.