I clicked the update button in the Hassio tab this morning and, after waiting an hour or so, noticed that I could no longer access the web interface. I have hard booted the Pi several times and nothing. SMB and SSH access still work, so I was able to grab everything in the config folder. My latest full backup was a month ago, however.
I tried renaming the database files and configuration file and did a restart. It automatically created new ones, but still not working.
Any ideas? Should I just blow away the SD card and start from fresh?
I thought of that and did switch the logger default to debug before another restart. The log is about 1200 lines long. Just post the entire thing here, or is there a specific section maybe? The lions share of the debug logs appear to be Chromecast I think (pychromecast.socket.client) “failed to connect” messages. I assume they are not related?
Looks good on first sight (quick look). No errors thrown by any component or core.
I wasn’t expecting a full log trace in this post. Be careful for sensitive data like your location etc. Just my 2 cents… That being said can you double check in your file (word search " ERROR ").
So far HA has never let me down when it comes to showing what is wrong in the logs. This leads me to believe that typically something else must be bad (network? computer?).
There are a couple of steps before you have to format your SD card (and loose your specific configurations like symlinks etc.). I’ve quickly written down what I would try in your case:
Quickly disable all components by commenting them quickly
If the UI starts: Re-enable them 1 by 1
a. Found the component blocking the start? Check what changed…
be. No success? Step 3
Try installing HA on other instance (by migrating the configuration YAML’s and other config files). Something that is very easily done when running HA with Docker.
Still no luck? Backup all your data… also those from extra apps you installed like Node-RED and more.
Fresh install
Copy paste everything back
Hopefully this will help you in finding the root cause.
I’m chasing down the network angle as nothing else is helping yet. I ran a netstat command on my RP3 and I don’t see port 8123 open at all. I’m not overly familiar with Linux, but is there a way to see if the web server is running? Anything else I should check?
I gave up this morning and started over. I installed a fresh hassio which worked fine. I then copied up my config files and reloaded components which worked as well. Everything was back up and running. I then went into the config to do some cleanup work - setting the recorder to purge after a few days - performed a HASS restart from the gui, and I’m back to the same issue. I can access via SSH or SMB, but not via the web interface. Sigh…
Had the same issue after upgrading from 0.81.6 to 0.82: no web interface. After the struggle with V3 Mosquitto broker this week, I think the best lesson is: wait a few days before upgrading
Just in case it helps someone else, I’m 90% certain that this is related to mqtt using mosquitto. I rebuilt from scratch three times today and would experience the same issue within an hour of installing add-ons. I then decided to simplify things and removed all of my mqtt code and skipped installing mosquitto. I have been running perfectly for about 6 hours now.
FYI
I had the same issue after an upgrade. Did some investigation, it turned out to be Google Chromecast that blocked it for me. Found some errors in the logging, disabled the discovery component, removed every cast device from the .storage/core.* files and it worked after a restart.
I also use mosquitto which is still in working order now.
It’s in your configuration.yaml. You need to add 2 lines to tell Home Assistant to NOT automatically add the chromecast devices. You should have the “discovery” line by default, so just add the 2 lines below it. Ensure that your spacing is correct:
# Discover some devices automatically
discovery:
ignore:
- google_cast
You may need to remove them from the “Known_Devices.yaml” file as well.