when the docker container starts it thus starts Neo4j itself. And when the docker container exits, Neo4j stops. There should be no read to 'start the db' by running /var/lib/neo4j/bin/neo4j start or similar
I started seeing that same error this week. I've rolled back to earlier versions of Neo and that did not help.
Wednesday: IT system updated my Virtual Box to version 6.0.10 (from 6.0.5 I believe). I run Ubuntu Xenial inside that, and then Docker, and then Neo4j.
Thursday: First noticed this problem, could not resolve the problem in Docker (clearing out volumes and images), so rebuilt the Ubuntu system from scratch. That resolved the problem.
Friday (today): Problem is back again. My development system was shutdown overnight, as normal.
So far, I got no idea .... but for sure, something is wrong.
We recently had the same issues on our multiple deployments.
We don't know the exact cause, but in our deployment scripts we were automatically upgrading to the latest last supported version of the kernel (centos). Disabling this role in our deployments resolved the issue.