If setup as Active/Passive on a single cluster node, the second server will pick up the slack. But what happens if the main server is fully functioning from a hardware perspective, but the database is corrupt or something along those lines?

Will it still failover? Or is there something else to setup for that?