This flag gets used when you suspect your copy of the blockchain may be corrupt, correct? What are the details?
It's a daemon flag? Is it supposed to be used just until the blockchain is synced, and then you exit and run with the usual flags you may use? Does this flag interfere with other flags? Under what circumstances should I run with this flag? Are there any other pertinent details regarding this flag?
If your daemon was interrupted by a power failure or OS crash, and it was still trying to catch up to the network (as opposed to already being synchronized), there's a chance the DB got corrupted. If you get a DB error after starting after a crash, you should try using this flag. You should only use this flag after you've already seen a DB error that causes monerod to fail. You don't have to exit and rerun or do anything else unusual in combination with this flag.
LMDB always maintains two snapshots of the database. This flag attempts to use the older snapshot instead of the newer one when it first opens the database, on the assumption that the newer one is the only one that got corrupted by a crash. It may not always be successful. Once the database has been opened and operated on successfully, the flag has no further effect.
Using this flag when there hasn't been any crash will erase the last transaction committed to the database, and make the daemon need to catch up to the network again.
