Human error is the "official" reason for a two-day outage of a web-based customer account management system used by America's second-largest long-distance phone company.

A spokesman for MCI's US parent company Verizon Communications told The Reg that "somebody simply ran the wrong process on the server" that's used by MCI's account management system. "In doing so, it prevented access to the web site." The explanation followed a call by this Reg reporter, unable to view or pay his bill for two days, to MCI's customer technical support team.

But, according to the techies we talked to, a database holding all the user names and passwords for MCI customers using the site was offline and not working following an upgrade designed to handle a growing number of customers signing up to MCI's online system.

Users trying to view and pay their bills were greeted with the message: "We have encountered a system error which makes us unable to fulfill your request."

Were the tech boys fobbing off customers with their high-falutin' tech talk or giving out more information that corporate would like?

MCI's Verizon parent insisted the matter was one of "site access" adding no customers' records had been removed and nor were they in danger of disappearing.

"The problem was discovered via the trouble desk and was reported to the individuals who were at the root of it. They re-ran the process, correctly, and problem solved," the spokesman said.®

Related stories

• Hello, this is Oracle - we're not in right now (29 August 2008)
• Gmail outage causes outrage (12 August 2008)
• 1&1 botches Microsoft Exchange update (8 August 2008)
• Apple says MobileMe mail problems 'behind us' (30 July 2008)
• Paypal glitch hits merchants with 12-day headache (27 May 2008)
• HSBC e-payments system goes titsup (again) (8 April 2008)
• Royal Mail sites hit by downtime cock-up (4 April 2008)
• Amazon's cloud now less prone to failure (27 March 2008)
• Update 3 Hotmail dies on both sides of the Atlantic (26 February 2008)