The Hemel Hempstead oil terminal fire has bought DR to the fore. We covered it thus:

- BakBone's replication product announced,
- my web log about the impracticability of SMB DR.

Now we find out how a software and outsourcing specialist called Northgate has been affected. It's not pretty reading. We might have assumed that an outsourcing specialist would have well-rehearsed and slick DR procedures with lost IT online time restricted to the hour or two needed for backup systems to switch over.

Northgate's HQ building was almost destroyed in the fire. Saturday's backup tapes were awaiting transport off-site and were destroyed. Data from Sunday was also lost. Northgate's clients' data will have to be 'recompiled.'

If I were a client of Northgate, such as one of the five hospitals mentioned, or one of the 90 percent of UK local government entities, Tesco, the Labour Party and Manchester United, I think I would be annoyed.

The company has set up new hardware services at its 25 other UK facilities and moved its HQ operation to a London office.

Well and good, but why didn't Northgate have a disaster recovery plan that anticipated the facility loss and had a rehearsed procedure to cope with it?

Apparently it did. Here is what its statement said: "The fabric of the (HQ) building and the fixtures and equipment inside have been badly damaged. The back-up systems that were in place have also been rendered inoperable. Northgate’s ability to service its customers has therefore been temporarily affected. As a result, its well developed and previously rehearsed business continuity plan has been initiated. This will allow Northgate to restore its services using remote data centres and the other offices that Northgate has around the country."

The backup systems are "inoperable". Northgate's "well developed and previously rehearsed business continuity plan has been initiated." This business continuity plan left data in rest at risk. It left backup tapes waiting to be shipped out and thus at risk. You might think that this is simply not good enough.

Northgate certainly didn't use its network facilities to transport the data off-site.

What does Sunbelt do? Ian Masters, Sunbelt's sales director, said: "Sunbelt System Software’s DRP utilises NSI Software’s Double-Take software to replicate Exchange and File data from our UK to our French office. We then also replicate the Exchange, File and SQL data from France to the UK, plus both offices backup all data to tape at the end of each day. This protects us from site failures and with the VPN technology that we have already implemented key members of staff can access critical data from home."

Northgate did not have this level of DR capability. Should it have? Its clients might think so.

We're talking about a firm with 3,000 employees and a £200 million plus turnover. Its disaster recovery plan has meant its customers have lost data and suffered interrupted servicve.

Here is what the company's web site states about customer data and risk: "We are passionate about quality and ruthless in our quest to continually develop lowest risk solutions that are easy to use and rich in functionality."

Three cheers for marketing-speak. Isn't it drivel?

Northgate recently won "a Six Sigma award for Excellence;" surely not for its disaster recovery and business continuity planning.

There is an obvious conflict between tendering for outsourcing contracts and winning on value for money terms, as well as on other factors, and on providing relatively expensive DR facilities.

If a £200 million IT outsourcing specialist doesn't have an effective disaster recovery plan then why should its customers have faith in its abilities as a managed service supplier?

If such a firm finds it too expensive to have proper DR facilities in place then why should any SMB business bother buying them?