Lovelace access KO but API access OK

Over the last few weeks, my HA (running on a debian VM, hassio installed using the generic linux installation) stops working properly.
The API access continue to work (eg https//hassio:8123/api/config is OK).
However, trying to access lovelace web ui gives this :

with the spinning wheel in the middle and I never gets the normal UI…

I have another instance of HA running on a RPI on my LAN. In order to troubleshoot this as soon as it happens, I would like to run some nodered flow on the RPI, accessing the web ui and warn me when something is wrong.

The problem is that I can’t make the difference using nodered between the web ui working and the spinning wheel. Both are giving on “200 OK” as HTTP response. I would need to do something more clever to distinguish between the “normal” UI and this wheel.

Any suggestion on a way to achieve that?

The next step will be to understand why my hassio instance is doing this… But that will be for later :slight_smile:


Since this post, it happened to me twice. Same symptoms. Is there a way to distinguish between the “normal” UI and this wheel programmatically? Any clue?

edit: try wiping browsercache first

It probably means frontant cant communicate with homeassistant

here’s a good start:

  default: debug

tail -f /config/home-assistant.log

when you found it specify your logger to something like:

  default: info
    homeassistant.components.frontend: debug
    homeassistant.core: fatal

afterwards build an automation for that kindof event:

  - alias: notification for "frontend" errors
      platform: event
      event_type: system_log_event
      condition: template
      value_template: '{{ "frontend" in }}'
      service: notify.pushover
        title: The wheel! It's spinning!
        message: '{{ }}'

Thanks for the tip. I have done that. Now, I’ll just have to wait some time…