Twitter fixed after outage – data center failure blamed

twitter-badAlmost a month to the day since its last major outage, microblogging site Twitter broke down again on Thursday, leaving millions around the world unable to tweet their thoughts/opinions/recommendations/grievances in 140 characters or less.

The site was out of action for around two hours, from about 8.20am PT. For a while, visitors to the site were met with the message: “Twitter is currently down for <%= reason %>. We expect to be back in <%= deadline %>.” Presumably at that point they had no idea what the reason was or when it thought it would be back, but for sure they didn’t want everyone to see the code like that.

It was Mazen Rawashdeh, Twitter’s VP of engineering, who once again took to the Twitter blog to offer an apology for the loss of service, beginning his post with the words, “We are sorry.”

He went on to explain just what happened: “The cause of today’s outage came from within our data centers. Data centers are designed to be redundant: when one system fails (as everything does at one time or another), a parallel system takes over. What was noteworthy about today’s outage was the coincidental failure of two parallel systems at nearly the same time.”

So it sounds like some parts of the Twitter infrastructure are creaking under the sheer weight of users. “We are investing aggressively in our systems to avoid this situation in the future,” Rawashdeh said in his post, adding “We apologize deeply for the interruption you had today. Now, back to making the service even better and more stable than ever.”

The outage comes just over a month after a ‘cascading bug’ was blamed for knocking out the service, again affecting users worldwide.

No doubt the San Francisco-based company is on guard to prevent the site from imploding on Friday in the face of a deluge of Olympic-related tweets expected during the opening ceremony of the sporting extravaganza in London.