Recorder restore_state not working after HA "crash"


when restarting my HA docker container (and not restarting HA with the frontend UI itself) I get the following error in the logfile:

‘2018-02-15 10:01:43 WARNING (Recorder) [homeassistant.components.recorder] Ended unfinished session (id=372 from 2018-02-15 09:00:20.525449)’

When I see this error, the restore_state function is not working as expected, all switches which were “on” before the restart are then “off”. This behaviour can be always reproduced by just unplug HA instead of restarting it gracefully with the frontend.

Ist this behaviour expected or a bug?

any help/hint to this? unfortunately thats a big problem for me, after each restart of my docker host (synology diskstation) I`ll have to check manually if all state are correct :frowning: