What to do/check after a HA crash

As with any system there are often considerations that need to (or at least should) be made after a crash.

Recently I have been experiencing crashes on my pi4/ssd setup. (Samsung SSD/the recommended dongle to the SSD and a 3.5A power supply)

I reviewed my automations and have found some orphaned references that I believe could have been part of the problem. (We’ll see)

My setup is <1 year old and I only have around 60 Insteon devices and around 50 automations. (not a big install I am guessing).

I have not yet opted for things like an alternative to the standard DB, etc. (IOW - it is currently a plain vanilla setup)

This post is NOT an ask for help debugging my setup - but rather an ask that when HA crashes, what should one look at from the SYSTEM aspect???

That is, can the DB become corrupted? (And is there a fix for this?)
Can the file system become corrupted?
Should I revert to a previous snapshot that wasn’t crashing? (Is that actually a NEED as in being able to restore non-corrupted objects?)

Any thoughts around this aspect of system maintenance are appreciated.

Thanks anyone who responds here…

1 Like


You can delete the db and it will rebuild it on next restart.

possibly, but less likely unless the system disk is corrupted and/or a update was interrupted.

If there is nothing in the home-assistant.log that gives a hint at the problem then that’s likely the only real recourse to fix it to a point that you can re-create the conditions that caused the failure so you can do some other thing or report an issue.

There are also topics on the forum about repairing the database if you are really attached to your short term data.