The recent IRA bombings have brought the peace process and it's implications back into the limelight and just about every magazine and it's dog have decided they must have an angle on it. I've seen various headlines in obscure and well known titles alike ranging from 'Pig farmers reel from Docklands bomb' to 'Peace process fears thwart light refreshments initiative'.
The IT industry sadly has succumbed to the same 'cashing in' syndrome with headlines like 'Bytes for Belfast as terror returns'. However the biggest con seems to be the fact that national and trade newspapers have latched on to the whole disaster recovery thing as the story that really gives the bombers one in the eye. I was able to keep myself under control until I saw this headline from a recent edition of a leading trade weekly (don't want to upset anyone you understand) as an example 'Disaster recovery' thwarts bomb attack' and just had to rant off.
The story goes on to tell us that a Swiss software firm was fully operational on Monday morning after its US, European and African subsidiaries were able to resume 'critical' business using portable PCs. The company's chairman claims that this fast response to the 'terror bombers' is purely down to regular tape and disk backups. The other company quoted in the story is Bell cable which was cleverly able to relocate it's network to a recovery site in North London within 20 minutes of the blast.
I must not be alone in thinking that what these companies have in common, is that they are IT companies ,and as such have either learned the hard way or have super techies in the house who get as much pleasure out of backup as Yorkshire men get out of pigeons, faggots and boring batsmen.
What I want to know is what would have happened if the company had been a typical UK user of computers, whose primary business was pork pie retailing, and whose offices had been anywhere but in docklands, which after all is supposed to be London's high tech citadel.
Most people know that a tape backup is a great peace of mind tool, but in reality it would be easier to retrieve a licorice allsort from a coal bunker. Yes you would find it in the end, by systematically going through all the coal, but you'd have wasted so much time doing it that you'd probably gone out of business anyway.
I'm probably being unfair to tape backup machines, but they are just so difficult to use. Usually the IT department install them, then gives you a box of photocopied 'how to' notes and reassuringly tells you you'll have no problem. You then spend a few weeks religiously backing up the machine, but the days get busier and your machine is so reliable anyway that the time between backups gets longer and longer.
Then guess what! The server bums out and that file you really want is on badly labled tape number three and has a really obscure file name like burp.xls, which you named that Friday afternoon, after your department had been out for a team curry. You however think it's called budget.xls because you always use generalised difficult to remember names and can't find it. Even if you could, you'd need a large box of tools, some jump leads and that guy from next door who's like Magnus Pike but much brighter.
You know I'm right! Just read the pages of horror in any disaster recovery firms product literature. It reads Like a Stephen King novel and the really frightening thing is that human error cocks it up every time. Computers they're fine, backup that's fine but procedures well we forgot about that sorry!
And distributed computing has made it all worse. Even if you've got to grips with structured back up and disaster contigency planning most of the critical systems and data are spread throughout an organisation. After the disaster, you can repair the damage at the data centre, but the company remains effectively dead in the water because critical systems and key data reside on departmental servers and thousands of PCs, that never get backed up and are stuffed full of files called burp, bum and botty. And the network that ties everything together remains in shambles.
What you need to do is make sure that every LAN and server is treated as a seperate data center, and that each has its own disaster plan, including hot-site arrangements and disaster-recovery procedures. That as you know is as likely as finding that licorice in under five minutes or remembering that burp is really budgets.
So don't tell me the bombers were thwarted by disaster recovery, all the firms involved were lucky. How would you have fared?
