“You are currently running version 0.108.9” I have been trying to update several times from 0.109.0 to 0.109.2 , update looks like its working, it goes offline for a few mins, and sometimes it’s down for a good 10++mins, once it comes backup I’m presented with the same upgrade card.
I’m running hassio on a docker instance.
Checked from Supervisor/System Log and I don’t see anything from there that would be causing it from updating.
1 Like
SSH in, and enter
ha core update --version 0.109.2
What do you get ?
I get the following after a while, I then try going back to the main Supervisor screen, the system is offline for several minutes. (maybe a good 10++ mins) before hassio comes back online. This time it didn’t I’m just going to reboot and let you know the status when it comes back up…
No joy after the reboot. Back to the upgrade screen.
I have exactly the same issue on my nuc using ha core update, I then used
ha core update --version 0.109.2
As @francisp told, my setup was offline for maybe 2 min and now it’s back up and running 
fully updated, thx!
What does the developer-tools/info page tell you?
Still showing " Home Assistant 0.108.9"
OK, a frequent reason for upgrades failing is lack of space. Do you have enough?
df -h
I will have a look shortly, but the HD on the system is 22% full , but I haven’t ssh’ed in hassio…
I was just looking at the Logviewer and noticed the following message for the database…
(Background on this error at: Error Messages — SQLAlchemy 2.0 Documentation)
2020-05-01 20:04:36 ERROR (Recorder) [homeassistant.components.recorder] Error executing query: This Session’s transaction has been rolled back due to a previous exception during flush. To begin a new transaction with this Session, first issue Session.rollback(). Original exception was: (sqlite3.DatabaseError) database disk image is malformed
[SQL: INSERT INTO events (event_type, event_data, origin, time_fired, created, context_id, context_user_id) VALUES (?, ?, ?, ?, ?, ?, ?)]
[parameters: (‘state_changed’, ‘{“entity_id”: “sensor.enderpro_job_percentage”, “old_state”: {“entity_id”: “sensor.enderpro_job_percentage”, “state”: “46.48”, “attributes”: {“unit_o … (531 characters truncated) … :00”, “last_updated”: “2020-05-02T00:04:35.298277+00:00”, “context”: {“id”: “5189824ef76b4db8ada5a2d91f2aebbc”, “parent_id”: null, “user_id”: null}}}’, ‘LOCAL’, ‘2020-05-02 00:04:35.298314’, ‘2020-05-02 00:04:35.304830’, ‘5189824ef76b4db8ada5a2d91f2aebbc’, None)]
(Background on this error at: Error Messages — SQLAlchemy 2.0 Documentation) (Background on this error at: Error Messages — SQLAlchemy 2.0 Documentation)
2020-05-01 20:04:36 ERROR (Recorder) [homeassistant.components.recorder] Error saving events: This Session’s transaction has been rolled back due to a previous exception during flush. To begin a new transaction with this Session, first issue Session.rollback(). Original exception was: (sqlite3.DatabaseError) database disk image is malformed
[SQL: INSERT INTO events (event_type, event_data, origin, time_fired, created, context_id, context_user_id) VALUES (?, ?, ?, ?, ?, ?, ?)]
[parameters: (‘state_changed’, ‘{“entity_id”: “sensor.enderpro_job_percentage”, “old_state”: {“entity_id”: “sensor.enderpro_job_percentage”, “state”: “46.48”, “attributes”: {“unit_o … (531 characters truncated) … :00”, “last_updated”: “2020-05-02T00:04:35.298277+00:00”, “context”: {“id”: “5189824ef76b4db8ada5a2d91f2aebbc”, “parent_id”: null, “user_id”: null}}}’, ‘LOCAL’, ‘2020-05-02 00:04:35.298314’, ‘2020-05-02 00:04:35.304830’, ‘5189824ef76b4db8ada5a2d91f2aebbc’, None)]
I have been experiencing the exact same issue for several days. I have tried updating from the UI and command line many times, disabling tons of integrations/components (anything new, all config includes, recorder, etc).
I have execd into the home assistant docker while it is running 0.109.2 and he checked home-assistant.log but there is nothing abnormal there. I have run exec with check config while 0.109.2 is running and it comes back clean. Seems like it is up and running for ~10 minutes, but I cannot access the frontend. Then inevitably it rolls back to 0.108.9.
I have ~20GB free on my SSD.
The one big thing I have been avoiding is that I recently updated to Ubuntu 20.04. Maybe I should restore my backup from 18.04…
edit: only interesting thing in the logs:
20-05-02 02:28:28 INFO (MainThread) [supervisor.homeassistant] Successful run Home Assistant 0.108.9
20-05-02 02:28:28 INFO (SyncWorker_15) [supervisor.docker.interface] Cleanup images: [‘homeassistant/qemux86-64-homeassistant:0.109.2’]
I finally got 0.109.2 running!
It appears to have either been Glances or Wyzesense. Will experiment and update.
Good luck @digiital !
not using either of those. What did you do? Remove them? or did you change something in the configuration yaml?
I was just updating ‘again’ and watch the log viewer to see if something pop’ed up and nothing.
Side note. I fixed the sqlite file so i’m no longer getting the errors I posted above.
I commented out the wyzesense and glances entries in my configuration.yaml. Maybe I commented out something else without remembering. I’ll report back if/when I figure it out.
Well guess I won’t be able to try that since I don’t have either of those in my yaml.
Do you have hdhomerun in your config?
I do. let me disable it.
YAHOO!!! That was it! … I removed and HDHOMERUN and now I’m at 0.109.2
Are you using the hdhomerun integration from HACS? If so I think I know where the problem lies. https://github.com/burnnat/ha-hdhomerun
saw some error messages on mine that the library .so was missing. I installed it manually early tonight, but that didn’t fix it, was only disabling it that allowed me to update,
You didn’t answer my question, but never mind.
My opinion about what is going on here:
- Having hdhomerun in configuration.yaml appears to stop HA starting properly. Comment that out from configuration.yaml (I am unsure whether this is due to the translations directory error below, or the fact that integrations are now supposed to be UI configurable only)
- The component has not been updated for 0.109 because it still uses the .translations/ directory, it now should be translations/ (the difference is the leading . )
- Change it
mv $CONFIG/custm_components/hdhomerun/.translations $CONFIG/custm_components/hdhomerun/translations
- restart home asssistant. At this point I was able to use the UI integration to find my hdhomerun device(s)