This happened to me a few days ago, but I assumed it was because my rig rebooted unexpectedly due to a power outage due to my foot being in the wrong place at the wrong time 
My main node just refused to sync and there was an error about the sqlite database (I wish I would have saved the exact error now). At first, I just deleted the corrupted blockchain_v1_mainnet.sqlite db file and it started resyncing, but with the blockchain well over 2gb now, my farmer wouldn’t be farming for hours (days?) at those rates.
Luckily I had another laptop configured to run the full node (with UPNP set to false) and it had synced as of a few days before this error. It hadn’t been running in a few days, but it still had almost the full blockchain on disk. I copied the blockchain_v1_mainnet.sqlite from this laptop to my main node and then restarted the client. It worked! It started syncing from a few days before and was caught up in a few hours.
This experience made me realize that maybe it is a good idea to have 2 full nodes running (with UPNP turned off on one of course) OR maybe we should be backing up the \db directory for no other reason than it takes a LONG time to resync if that gets corrupted?