For the second morning, I've woke up to find my Hubitat isn't working, due to a corrupt database.
It was easy enough to fix. I performed a soft reset, and restored from a backup from the day before. Everything is back up and running. The only issue is, it happened again this morning.
I can't think of what exactly is causing it, but I suspect it's a specific "light wake" rule I set up in rule machine. I ran it once successfully on Monday and Tuesday, but made adjustments to it Wednesday night before bed. I just removed it entirely to see if that solves the problem.
Anyone else having similar issues? I have applied the last few platform updates as soon as possible, and I'm up to date.
Unusual for a rule to corrupt the database. Couple of things you may want to try just to rule them out:
- Do a manual backup
- Do a manual reboot
Both should complete without incident, but...
I'm not sure how likely it would be for a rule corrupt the database... I wonder if maybe either the problem is something other than a corrupt database or something other than the rule is causing the issue. I have had my hub go south when memory gets low, CPU spikes, etc... sometimes it is a corrupt database but often something else.
I use @thebearmay's hub info driver and a couple simple rules to capture critical data when the hub has issues... if you haven't looked at it it's super helpful.
To add on to @brad5's comment, sometimes a failing device will spam the hub and cause it to become unresponsive also.
How are you determining it is a corrupt database?
If you go to Logs tab, are there any errors?
In the Logs tab go over to Device Stats, and App Stats. Open the settings and check all the boxes. Are there any devices or apps that are consuming memory?
I would also install @thebearmay excellent Hub Information Driver and watch the memory usage.