Microsoft has apologized after problems left many users unable to access services such as Hotmail. Though it initially appeared the services themselves were down, it appears the problem was instead in the way Microsoft systems recognize users.
Some users were unable to access Hotmail, Windows Live Messenger and even some Xbox live services for around an hour.
Microsoft later put out a statement saying: “This morning, around 9:30 a.m. PST, the Windows Live ID sign-in service experienced a partial outage that caused some customers to not be able to sign into services using Windows Live ID for approximately one hour. The service is now restored to normal. Microsoft apologizes for any inconvenience this has caused customers.”
It followed this later in the day by explaining that one server used for processing log-in attempts had gone down, which caused problems when it increased the load on other servers. The firm then fixed
The situation is an interesting insight into the way firms prepare for such problems. By building enough redundancy into a system, it should be possible for a company the size of Microsoft to cope without any disruption if a single server goes out of action. That, however, would be more expensive and could be seen as an unnecessary expense. For a free service such as Hotmail, an hour’s disruption is arguably a price worth paying to avoid such unnecessary expense. But if Microsoft wants to get more businesses using internet-based “cloud computing” services for a fee, such disruption may not be acceptable.
The problems were also a sign of the benefits of sites such as Twitter. It certainly appears many users were able to discover from Twitter posts that the problem was widespread rather than their own accounts, which may have eased the number of queries to Microsoft and even cut down the number of attempted log-ins during the outage.