it’s even worse in the 0.81.0b0
Out of interest, what DB are you all running? I’m just getting ready to upgrade from 0.80.1 to 0.80.3 I use the MariaDB addon as it made things faster in the past. Just wondering if this is impacting all DB types or just a subset.
1 Like
I am running the default sqlite database.
I confirm I have similar issues- history graph loads „sometimes” since the 0.80->0.80.3 update.
Cannot find anything Interesting in logs.
Using MariaDB on external host, dropping DB did not help.
Tried also the local SQLite with a new file, same negative result.
Downgrading to 0.80.1 helped.
I upgraded to 0.80.3 yesterday, advising to postpone your update.
(Used SQLite then switched to MariaDB to get the charts queries)
Issue https://github.com/home-assistant/home-assistant-polymer/issues/1829 refers please comment there
Linked topics
1 Like
For me the issue also remained in 0.80.3. Rolling back to 0.80.1 solved the issue for me so far…
Has this been fixed in 0.81.0?
Yes. (Well I hope I haven’t spoken to soon!)
1 Like
So that’s one yes and one no. Guess I’ll have to try.
I normally don’t do x.0 updates but this time I think I will. I’ve already added entity_ids to all my template sensors.
well there’s a nasty lovelace bug with 0.81 so I’d wait for 0.81.1 if I was you…
I’m not using Lovelace yet.
This update has improved the graph display times for me. For the first page refresh the graphs are now actually being displayed after a delay of up to 10 seconds or so as opposed to not at all. Subsequent displaying of the graphs is quite snappy (< 1 sec).
My log is being flooded with these errors though:
2018-10-28 13:34:45 ERROR (SyncWorker_11) [homeassistant.components.recorder.util] Error executing query: (sqlite3.OperationalError) database is locked (Background on this error at: Error Messages — SQLAlchemy 2.0 Documentation
thats a database delete situation
Bear in mind that I am using a NUC Intel® Core™ i5-5250U CPU @ 1.60GHz as opposed to @DavidFW1960’s (I think) raspberry pi.
Yeah I figured as much. I was watching the other db files (.db-shm and .db-wal) they are being deleted after writes but the problem persists. Nuking the database is my next option but I’m about to go out and won’t have time to re-enter all my datetime inputs that get cleared. Will do it later this arvo.
I’m dying for a HassOS NUC image to be released.
Nick - I also run a NUC7CJYH but you outgun me with an i5 lol!
That’s what I do. Debian. I would never run HassOS on a NUC - way too restrictive for no gain at all.