Your comment reminds me of a news report here in Australia - 3 banks suffered a multiple-day outage when their IBM mainframes hosting transactional data were co

Your comment reminds me of a news report here in Australia - 3 banks suffered a multiple-day outage when their IBM mainframes hosting transactional data were corrupted so badly, the banks weren't sure they could even recover the transactions: http://www.smh.com.au/it-pro/business-it/the-real-reason-st-george-bank-of-melbourne-and-banksa-are-suffering-a-long-internet-banking-outage-20151005-gk1g62 http://www.smh.com.au/it-pro/business-it/the-real-reason-st-... The relevant paragraphs are these: "But when the system came back up early Sunday morning, a major data corruption was found by IT experts in the IBM mainframe Information Management System (IMS) database hosting the data, so all systems were shut back down about midday Sunday AEDT, a source close to the bank told Fairfax Media. "IBM IMS and CSC Hogan experts are trying to figure out how to restore the databases without losing any transactions that were done on Sunday morning," the source said on Monday afternoon, referring to IT outsourcers IBM and CSC being called in to assist, and Hogan, an integrated suite of core banking applications."