WorkCover retrieves data despite hiccup

A server outage at the WorkCover NSW resulting in a temporary data loss has been rectified with more than 98 percent of the information retrieved, according to the authority.

A WorkCover spokesperson confirmed to Computerworld that the Workers Compensation Commission server, located in Sydney, experienced a failure on March 27 but it did not affect any data held by WorkCover on workers compensation claims.

"The server failure did not affect WorkCover's main computer centre at Gosford (one hour north of Sydney)," the spokesperson said, adding there is no need for people to re-submit workers compensation claims.

"Over 98 percent of the data lost was retrieved within nine working days," he said. "The remaining less than 2 percent is being retrieved on a case-by-case needs basis."

The data was contained on the system's S: and H: drives, used for "common administrative, procedures, and correspondence" and "personal" information, respectively.

The exact cause of the server failure, or the data backup tools and procedures used, were not revealed by the authority.

IDC's Asia Pacific storage associate vice president for Graham Penn was blunt about the glitch, saying "it shouldn't have happened".

"At least 50 percent of organizations' data is not adequately protected [and] in the event of a major disaster you may not get it back," Penn said. "I'm sorry for them, but every organization should have systems robust enough to withstand a failure. It's a wakeup call for everyone."

Penn said any significant site has had more than adequate warnings that you need to test and test again your backup and recovery procedures.

"Failure to do so results in more costs and loss of time and reputation," he said. "I don't care what the failure was - it all should be included in backup and recovery."

With business-critical data, Penn recommends a semi-automated recovery period of two days be in place, because the "real cost" is between five and 10 times the IT cost in lost time and productivity.

"[Disaster recovery] should be tested every three months, or if it is really business-critical, every month," he said. "If it is that business-critical, you should have it switch over to an alternative site."

Penn said if WorkCover's main data centre is in Gosford, there should be a full working copy in Sydney.

"If Gosford is the live data centre it should be kept fully up to date all the time," he said. "Contingency planning should have been provided for [because] all hardware will fail at one time or another."

Join the newsletter!

Error: Please check your email address.

More about IDC AustraliaWorkcover

Show Comments

Market Place