"This wasn’t due to a hack or our new office or Euro 2012 or GIF avatars, as some have speculated today," Mazen Rawashdeh, Twitter's vice president of engineering, wrote in a post to the company's official blog.
Earlier Thursday, a group of hackers known as UGNazi said it was the force behind the outages, which were Twitter's longest in eight months. In an email to Reuters, UGNazi said it targeted Twitter with a distributed denial-of-service (DDoS) attack because of the company's support for the Cyber Intelligence Sharing and Protection Act.
But expanding on a tweet from earlier Thursday -- which spawned a parody account that was repeatedly suspended and restored -- Twitter blamed the outages on a "cascading bug in one of our infrastructure components."
A cascading bug, Rawashdeh writes, is bug with an effect that "cascades" into multiple software elements rather than remaining confined to one in particular.
Twitter engineers first discovered its desktop site was inaccessible and mobile clients weren't updating at about 9 a.m. Pacific Time. Service was restored just over an hour later before dropping again at about 10:40 a.m. for approximately 30 more minutes.
"We are currently conducting a comprehensive review to ensure that we can avoid this chain of events in the future," Rawashdeh writes.
Typically, Twitter says, its service is available for approximately 23 hours, 59 minutes and 40 seconds in a 24-hour period.