The outage started at 7:30 a.m. ET, with Eliason continuously issuing updates to his Comcastcares Twitter page throughout the day, several times forecasting fixes that failed to materialize.
Aside from Eliason’s Twitter messages, or Tweets, there was scant information coming from Comcast about the outage. A message to users clicking on Comcast.net’s Live Chat feature on its Help page apologized for the blockage and said, “At this time there is a known issue with our email system, our technicians are working as quickly as possible to resolve the issue.”
Apologizing for the hassle, Eliason said Saturday morning: “Multiple servers down after power issue.” After a false restart, with servers coming back up and then e-mail again being halted, Eliason kept up a constant flow of Tweets, in response to user questions.
“Just received notice that others that were fixed are being impacted again. I am sorry for this trouble and agree unacceptable,” he tweeted about six hours into the outage.
Mail appeared to be flowing again for most subscribers late afternoon, with e-mail being routed to some users more slowly than others. Some users appeared to still be waiting for old e-mail to come through 12 hours after the outage was reported.
Eliason, who according to the Comcastcares page Saturday had more than 14,000 active Twitter followers, has been widely credited with spearheading use of Twitter for corporate communications.
Comcast officials contacted as e-mail started flowing again were unable to immediately say how many subscribers were affected. Comcast had about 15 million high-speed Internet subscribers at the end of 2008, according to the company’s
“The issue impacted some customers’ ability to access their emails and address books, but did not affect their ability to access the Internet,” according to a company statement issued after e-mail started flowing again early Saturday evening. “It’s important to point out that no customer emails were lost in this process,” the statement said.
Minor e-mail outages are fairly common, but those that last more than a few hours are typically, depending on the number users affected, considered major occurrences at large service providers.
In March Google suffered a widely noted outage that affected many users globally and lasted for two-and-a-half hours.