Hassio upgrade from 0.104.3 to 0.105 fail

What integrations did you have?

I get this error when updating-

20-02-06 16:21:07 ERROR (MainThread) [hassio.api.proxy] Client error on WebSocke                                                                                                                                                             t API Cannot connect to host 172.30.32.1:8123 ssl:False [Connection refused].

What does it mean? How do I fix it?

mqtt with zigbee2mqtt and that’s all it’s quite a new installation

Got it working in the end, observed the restart after installing 0.105.2 getting stuck at udevd[10]: Starting eudev on the console so I removed the Google Cast integration as suggested here https://github.com/home-assistant/supervisor/issues/859 and was able to complete the first boot after the update.

This isn’t listed as a breaking change. I’ll try this next.

It didn’t work for me.

with the “Check home assistant” add-on do you have this error?

Log
starting version 3.2.8
[16:37:40] INFO: Don't worry, this temporary installation is not overwriting your current one.
[16:37:40] INFO: Installing Home Assistant: latest...
[16:37:40] INFO: Please be patient, this might take a few minutes...
Fatal Python error: initsite: Failed to import the site module
Traceback (most recent call last):
  File "/usr/local/lib/python3.7/site.py", line 73, in <module>
    import os
  File "/usr/local/lib/python3.7/os.py", line 27, in <module>
    import stat as st
ModuleNotFoundError: No module named 'stat'
[16:37:40] ERROR: An error occurred while installing Home Assistant:

Has someone found a solution?

That worked for me!!

same for me … once I commented the netatmo configuration it worked with no issue :slight_smile:

not worked for me,
i removed google cast , Netatmo, group, zone, but nothing,

that’s odd …

have you tried to create a new env from scratch? maybe would be good to start this way and start activating items step by step. I had to do that when I upgraded to 0.104 :frowning:

the good thing, I was able to clean up a lot my configuration :slight_smile:

My lovelace UI cards are all messed up, I don’t know wth is going on.

I have no netatmo devices, but I had to add this bloc to configuration.yaml to enable startup :slight_smile:

netatmo:
  client_id: dummy
  client_secret: dummy

So it’s solved for me

Are you saying that you had no netatmo devices and added this to the configuration yaml file fixed your not issue?

i have much trouble with this relese
because is impossible for me make thi upgrade , i make a new installation directly with 105, but. in this case i have much trouble to configure remote access because take over one hour for go online, and if i make any change in yaml al restart non riesce a partire piu.

Are there any updates? I’m still having these exact same issues… Don’t have netatmo running or anything, recently did a fresh install of 104.3, so not looking to do it again for 105.5…

Holy shit, this actually worked. I have no netatmo devices, but adding this code to my configuration.yaml, rebooting, and then performing the update actually got me to 105.5!

Worked for me too.
I have a netatmo weather station. I had to change the configuration.yaml as described, restarted HA and then the upgrade worked! Re-enable netatmo via integrations and everything seems to run fine.

I have been playing around for several weeks and finally got .104.3 updated to .105.5. I’ll list my steps to hopefully help others.

  1. Cloned SD card. This way it is very easy to go back.
  2. Updated to .105.5
  3. Lovelace fails to load but Samba and SSH work.
  4. Removed all my custom setting except my duck dns from configuration.yaml. Basically just the default plus the http address.
  5. Deleted the lovelace file.
  6. Restarted
  7. Lovelace loaded.
  8. Added sections of my configuration.yaml file one at a time, saved and restarted.
    When all configuration restored started on the lovelace
  9. Added all the community integrations Rebooted.
  10. Slowly added cards back in. Backing up lovelace and rebooting after each. This was a lesson learned as I added something wrong and after the reboot, Lovelace failed loading again.

I believe my error is improper formatting somewhere in Lovelace. I have a logger setting for hacs and all my custom cards and no errors are showing.

I didn’t change anything and it didn’t cause an issue in the past, but something does now. Maybe time to go to yaml for lovelace. To me the visual in the raw editor, the card editor and Json is inconsistent and confusing and leads to errors.

Hope this can help others.