Problem:
When you try to start a specific Docker container using the Synology NAS GUI, the container is being stopped unexpectedly and you see an error message like this in the logs:
Start container mycontainer failed: {"message":"failed to initialize logging driver: database is locked"}.
Signal container mycontainer failed: {"message":"Cannot kill container: mycontainer: Container 5136ddceeb46004c5b18f04eb9ec10cac3808938515874fc31185b0964232201 is not running"}.
Solution:
I fixed this problem by stopping the container, duplicating the container session: Right click on the container -> Settings -> Duplicate Settings
That will create a new container with the given settings. Note that local ports will be set to Auto and will not be copied over, so if you use fixed local ports, you need to set them to a different value in the original container and then set the local ports on the new container to the desired fixed value. Also note that files inside the container are not copied over. In my configuration, all relevant files are stored in mapped volumes on the NAS.
The root cause of this issue seems to be that the logging database for this specific container has been locked by some process. The issue is always limited to a certain container and will not affect other containers (though it could in principle occur for more than one container). I know that at least in my specific case, the issue is not caused by a reboot and will also not be fixed by a reboot of the Synology NAS. Just before I encountered the issue, my NAS had not been rebooted for months, but it might be related to Synology package updates since I updated some packages using the Package manager just before encountering the issue, including a Synology Mail Plus update which failed on the first attempt, but succeeded when I clicked Update again.