This is what i have:
pi@hassbian:~$ cd /srv/homeassistant
pi@hassbian:/srv/homeassistant$ ls
bin hassbian include lib pyvenv.cfg share
pi@hassbian:/srv/homeassistant$
pi@hassbian:/srv$ ls
appdaemon homeassistant homeassistant_3.5.3
pi@hassbian:/srv$
Looks like my instructions are probably safe then. Worst case, if it breaks. Just “mv homeassistant-old homeassistant” and you’re back to the way it was.
Ok as soon as i have some time i try this and let you know… thanks a lot for now
Just last question… How to become root when in pi@hassbian: ?
It looks like you may not need to be root. This command should give you the access you need.
Ok in my case the problem was a scene in scene.yaml, that was referring to an entitiy group.ceiling_lihts that does not exist.
Deleting that entry in the scene.yaml file fixed the issue
How you checked the configurations to find that error?
EDIT: I checked the configs with hassbian-config scripts and all returned OK, so i don’t have any errors in my configurations…
while in the 0.101.3 (which was not loading the frontend) I ran, while in core-ssh
hassio ha check
but you don’t have hassio, so you need other command (maybe hass --script check_config) , make sure you run it with the new version …
Disabling api-password did the trick for me with my difficulties when upgrading from 0.100.3 to 0.101.3: This was the solution (hard to figure out though):
I tried also this but without success… Now I will disable also everything related to the scenes domain and try again. Why you disabled also the base_url?
Well I disabled both and I cant see any difference - but I will try to investigate what it is used for.
I did it following your indications, but, first of all i was forced to use sudo for all commands as using “sudo su -s /bin/bash homeassistant” i got the perimssion denied. Then after using the command “python3 -m pip install homeassistant” the system downloaded all packages but at the end i got:
Installing collected packages: text-unidecode, python-slugify, pyyaml, attrs, voluptuous, pytz, astral, PyJWT, chardet, idna, urllib3, certifi, requests, async-timeout, pycparser, cffi, six, cryptography, ruamel.yaml, MarkupSafe, jinja2, multidict, yarl, aiohttp, voluptuous-serialize, more-itertools, zipp, importlib-metadata, bcrypt, homeassistant
Could not install packages due to an EnvironmentError: [Errno 13] Permesso negato: '/srv/homeassistant/lib/python3.7/site-packages/text_unidecode-1.3.dist-info'
Consider using the `--user` option or check the permissions.
What to do now?
EDIT: used sudo also for installing homeassistant, it loaded all dependencies and at the end i got.
Loaded: loaded (/etc/systemd/system/[email protected]; enabled; vendor preset: enabled)
Active: failed (Result: exit-code) since Wed 2019-11-13 16:04:14 CET; 29s ago
Process: 7333 ExecStart=/srv/homeassistant/bin/hass (code=exited, status=203/EXEC)
Main PID: 7333 (code=exited, status=203/EXEC)
nov 13 16:04:14 hassbian systemd[1]: Started Home Assistant.
nov 13 16:04:14 hassbian systemd[7333]: [email protected]: Failed to execute command: No such file or directory
nov 13 16:04:14 hassbian systemd[7333]: [email protected]: Failed at step EXEC spawning /srv/homeassistant/bin/hass: No such file or directory
nov 13 16:04:14 hassbian systemd[1]: [email protected]: Main process exited, code=exited, status=203/EXEC
nov 13 16:04:14 hassbian systemd[1]: [email protected]: Failed with result 'exit-code'.
I am completely out right now… What to do?
Did you make the backup of the locations like he spelled out? All you have to do is delete the created folder and start again. And if you want to go back, you move the working venv folder back. Do you understand folder structures and what the commands are doing?
Yes i already did it to have 0.100.3 running again…
What do you mean to start again? Where were the errors i did in the first try?
I followed indications that @swiftlyfalling gave to me, but maybe i was wrong in something not told…
These statements contradict each other… My comments were based on
This line here tells me that, most likely, the pip3 install homeassistant wasn’t done in the venv.
I.E. step 8 either wasn’t done or executed with an error.
Another option, however, is to backup your homeassistant config and just reinstall everything from the very beginning. Then restore your config. Everything HomeAssistant needs is in the config directory.
I think you had issues with permissions because /srv is likely not owned by the user I thought it was.
Well i followed your instructions step by step, but i began all being pi user.
I tried to be root using your indication but it failed as i was asked to run with sudo command to be allowed.
Now i am back to the old installation.
The /srv actually is owned by root while /srv/homeassistant is owned by homeassistant.
So please tell me again, knowing now more infos, step by step what i must do.
I discovered also that the new /srv/homeassistant directory doesn’t have the hassbian directory which is present in what i am running now. Eventually do i need to copy to the new homeassistant directory?
To be more helpful this is my directory tree with 0.100.3:
And this is the one created with your instructions:
Hope this help more than my english…
Sorry my bad english, was i was telling is that i was back again running good HA. My question was about what to do to recreate the new venv as the first try was unsuccesfull.
Just to minimize confusion and keep everything being suggested in the same thread I’m going to re-post what I suggested in the other thread here also:
https://community.home-assistant.io/t/0-101-airly-apprise-sinch-solar-log-microsoft-teams/145370/214?u=finity
However, just to be explicit and using a “shotgun” approach try this…
rename your existing “configuration.yaml” file to “configuration.old”. restart HA. At that point HA will (should…) recreate a default configuration.yaml file (your HA frontend will look way screwed up but don’t be alarmed at that).
Then update to v101.x.
if you don’t get any more errors then at that point just start adding back pieces of your original configuration from the configuration.old file one section/domain at a time and restarting HA after every change.
Then when HA starts getting errors again the you know that the last change was a problem. Then you can try to fix that section from there. Once it’s fixed then continue adding sections one at a time until they are all added back in with no errors or are fixed (there could be more than one section that is causing errors).
I’m assuming you have been updating HA in the same way going from v100 to v 101 as you have done the updates in the past and it has always worked then. Is that correct?
