Re: Bank has Crashed
- From: Jim Smith <jim@xxxxxxxxxxxxxxxxxxx>
- To: oracle-l@xxxxxxxxxxxxx
- Date: Sun, 15 Nov 2009 11:15:40 +0000
In message <7765c8970911140425x4f39fcaak34172b11ba19865c@xxxxxxxxxxxxxx>, Niall Litchfield <niall.litchfield@xxxxxxxxx> writes
When I worked for the Halifax a few years ago, they had quite extensive DR capability with dark fibre replication between two data centres. One in halifax and one in bradford. In theory, operation was site neutral and all internet facing systems were supposed to be active-active across the two sites.Hi Looks to me from the outside to be rather a case of not thinking carefully enough about what happens if the entire datacentre goes. Not just for core bank databases, but all the other 'non-critical' systems (I'd lay a pound to a penny that customer appointments was classed as non-critical). Personally by the way I think a non-IT solution is fine, it's what works and so long as people can go to a branch and get cash then that would seem to be fine. On Sat, Nov 14, 2009 at 12:10 PM, Howard Latham < howard.latham@xxxxxxxxx> wrote: Our National; Halifax bank network has crashed for several hours now (Power Failure!) - Any one know what database they used - And did they have a DR Plan? As far as I know the plan is pen + paper. And ring customers who have appointments BUT they cant get the Phone numbers as they are on their computer system! -- Howard A. Latham
The core banking system is an IBM mainframe with lots of SQL Server and a few Unix oracle satellite systems. I suspect Niall is right and they neglected to keep the satellite systems' DR capabilities up to speed.
-- Jim Smith Ponder Stibbons Limited 251 Barcombe Avenue London SW2 3BH -- http://www.freelists.org/webpage/oracle-l