The software giant has 360 million Windows Live Hotmail users globally as well as 299 million active Windows Live Messenger accounts.

Though Microsoft thought it had restored service for all Hotmail, Office365 and SkyDrive accounts by 9:45 PM PT Thursday, some customers continued to report problems. At 11:02 PM, the software giant said it was working on propagating the requisite DNS configuration changes for fixing the problem, but that it would take some time to restore service to everyone.

Microsoft notified Windows Live users at 11:49 PM that it had completed propagating its DNS configuration changes around the world, and had restored service for most customers.

"Depending on your location you may still experience issues over the next 30 minutes as the [DNS] changes make their way through the network ," explained Microsoft blogger Chris Jones. "Thank you for your patience as we have worked to address these issues."

Lost In Translation

DNS translates Internet domain names meaningful to human beings into the numerical identifiers required by networking equipment to locate and address computing devices worldwide. Moreover, the system makes it possible for domain names to be assigned to groups of Internet resources and users in ways that are independent of each entity's physical location.

DNS is often compared to a telephone book because it likewise cross references individual names with their numerical equivalents. For example, the domain name cnn.com has an equivalent IP address of 157.166.224.25.

As Thursday's Windows Live outage indicated, however, DNS issues can have unintended consequences for users of services that are based in the cloud. What's more, fixing an unexpected DNS problem takes time. On Friday morning, Microsoft was still working to resolve the continuing problems experienced by a small number of Windows Live users, according to an Office365 tweet.

Nothing is 100 Percent Reliable

This week's Windows Live service disruption followed on the heels of an August outage that likewise prevented some Office365 and SkyDrive users from accessing their accounts for several hours. Those Office365 customers hit last month by the outage are slated to receive "a 25 percent credit on a future monthly invoice," Microsoft said in a statement given to the Seattle Times.

However, Microsoft's latest cloud difficulty was overshadowed in Southern California, where 1.4 million residents of San Diego County and adjacent areas were without electrical power from Thursday afternoon into early Friday morning. The electrical grid shutdown, which also affected parts of Arizona and Mexico, was being investigated by U.S. regulatory authorities.

The juxtaposition of the two service disruption events demonstrates that cloud environments are not the only infrastructures vulnerable to outages from time to time.

"Nothing will ever be 100 percent reliable and -- especially in IT -- nothing is going to stay the same forever or even for a long enough time," said Andrea DiMaio, vice president and distinguished analyst at Gartner Research.

"So, besides learning to feel less desperate if tweets are stuck or VOIP does not work, we all need to really understand that technology is fallible, and we will always need to exercise our creativity to react to unexpected situations," DiMaio wrote in a blog late last year.

DaveB:

Kel C.:

Linda:

Goldie:

Carolyn P: