Problems after last upgrade

Upgraded earlier today to the latest version through the web interface like I have done every other time. Strangely enough this time was that I couldn’t access it afterwards.

  • Tried restarting and no improvement.
  • Nothing happens on my duckdns ip or the internal one.
  • The device boots up, red light on my RPI and occasional yellow indicating activity.
  • The device appears in my Google Wifi device list and I can ping both hassio.local and the actual IP.
  • Ssh turned off unfortunately and no backups.
  • When pluging in the SD card in my mac I get “not readably by this computer”

Quite desperate now and any suggestions on what to do?

Your typical PC won’t be able to read the format of the SD card. Try connecting the HDMI output of your Pi to a monitor or TV and watch it boot up.

Thanks. Little more information in image below but unsure what to make of it:

Did actually managed to get into the SSH:

core-ssh:~# hassio ha info
{
    "result": "ok",
    "data": {
        "version": "0.83.2",
        "last_version": "0.83.3",
        "machine": "raspberrypi3",
        "image": "homeassistant/raspberrypi3-homeassistant",
        "custom": false,
        "boot": true,
        "port": 8123,
        "ssl": true,
        "watchdog": true,
        "wait_boot": 600
    }
}

And more importantly:

Error decoding json invalid character '-' after top-level value: 2018-12-12 21:26:26 ERROR (SyncWorker_0) [homeassistant.util.yaml] while scanning for the next token found character '\t' that cannot start any token in "/config/scripts.yaml", line 683, column 1
2018-12-12 21:26:26 ERROR (MainThread) [homeassistant.bootstrap] Error loading /config/configuration.yaml: while scanning for the next token found character '\t' that cannot start any token in "/config/scripts.yaml", line 683, column 1
2018-12-12 21:26:26 INFO (MainThread) [homeassistant.core] Starting Home Assistant
2018-12-12 21:26:26 INFO (MainThread) [homeassistant.core] Timer:starting

Any idea what is going on? Is it config.yaml related? Strange since I didn’t change anything there if so…

Hi,

Just tested this tonight because of the same problem.
Used command from SSH : hassio ha update
And the problem was gone !

First the file looked like this:

core-ssh:~# hassio ha info
{
“result”: “ok”,
“data”: {
“version”: “0.83.3”,
“last_version”: “0.84.1”,

Why is version lower than last_version ??

After update it changed to this:

core-ssh:~# hassio ha info
{
“result”: “ok”,
“data”: {
“version”: “0.84.1”,
“last_version”: “0.84.1”,

Regards Gunnar.

1 Like

Thanks.

Tried it out and eventually got it working. Now when checking there I have the correct version number as well :slight_smile:

1 Like

I have a different problem. Ran the update from hass.io. Now the Overview panel displays normally, but the hass.io panel shows an error:
image
It gets worse. All the other panels show this:
Capture2
A quick SSH into HASS.IO suggests everything is started and running, but perhaps just not displaying. I am at a loss what to do!

Mess. Had problems as well since so think I will just do a clean install after holidays :frowning:

1 Like

I think I will too. Things are a bit worse than I thought. My automations have also stopped working. It occurred to me I can at least copy the “configuration.yaml” file, (and perhaps others?) so I don’t have to rewrite it from scratch when I reinstall. Any tips from people out there who have been down this road before would be most appreciated!

Agreed - mess !
I upgraded via the UI to 84.2 from 82.1.
Interface starts, and appears to be running - but none of my devices are found. Overview page is blank.
Cant SSH to host (connection refused) now.
I feel a fresh install coming on - just got to dig out config files and scripts etc…

First, go into each of your automations and add right after your alias line:

initial_state: true

Next, go to your sensors and everywhere you have a template, add the entity_id: line referencing the sensor that updates your template. i.e.:

- platform: template
  sensors:
    spacex_next_launch:
      value_template: "{{ states.sensor.spacex.state | int | timestamp_custom('%a, %b %d at %I:%M %p')}}"
      entity_id: sensor.spacex

Then, (if you used owntracks) lose your old owntracks stuff in device_tracker: and replace it in your configuration.yaml with:

owntracks:

Then reboot and pray.

1 Like

Thanks, Dixey. Will attempt.
PS I use GPSLogger, not OwnTracks. Do you think similar advice about device_tracker would apply?

I have not seen any issues or changes with GPSLogger.

Mine did the same thing after hitting the update button. No access to web user panal.
Can ssh into root
core-ssh:~# hassio ha update
ok
core-ssh:~# hassio ha info

Edit:
** Problem was unrelated to the upgrade. I had been editing and entered Location in wrong format, and then did the update without restarting. The bad location borked it. Edited configuration.yaml with correct location lat/lon format in samba, now all OK.**

I saw ipv6 errors, I usually have ipv6 disabled in my network. Once I enabled IPV6 in my router and rebooted all my Sonoff lights turned on.

Thanks,

It seems my problem disappeared while I was at work. I’ve checked the log, and a series of ‘TCP errors’ stopped this afternoon. I’ve read in other places those errors may be related to Home Assistant connecting with the Chromecast device in my house. Interestingly, now that it’s working my Google Home Mini and Apple TV don’t appear as media players in the Overview screen anymore (which in my case is not a concern).
I wish I was smart enough to figure out what happened!

Glad to hear you are no longer dead. Now, on media players disappearing… my experience has been that if a media player is physically powered down/turned off/unavailable during the restart of HA, the media player entity may not rebuild unless it is automatically discovered.

Thanks, good to know.

I did a clean install earlier today. fresh from the hassio link.
its still dead.
As others have reported several of the HA UI pages don’t load - or appear blank.
The web UI will report ‘reconnecting’ for no apparent reason at certain times. Even with the only add-on installed is SSH.
All my ORVIBO devices are no longer recognised. and my cards don’t appear.

I notice that customize, scripts, and groups.yaml all have zero sizes in the config folder.