Recover Option 1

1.	Correct hardware problem (such as the downed network).

2.	Start the run again in recover mode which causes each server in each cluster to pick up where it left off.