Leap second and storm disrupt weekend web services
Several high-profile websites and services suffered over the weekend thanks to the leap second and huge storms.
An extra second added to the world clock saw the likes of Reddit and Foursquare fall over.
Hours earlier, a lightning storm on the American east coast left millions without power, and disrupted a major cloud storage data centre.
Netflix and Instagram were among the services affected.
Amazon Web Services (AWS) - the company's cloud computing servers - was in part taken out by the storm, as was its back-up generator.
Amazon's service is used by hundreds of companies as a way of cheaply hosting content online.
"We will share more details on this event in the coming days," Amazon said in a statement.Every second counts
As websites recovered from the storm-related disruption, more trouble was on the way for web users thanks to the leap second - an extra second added to the world clock to compensate for irregularities in the earth's rotation.
On 30 June, at 23:59:60 (00:59:60 BST), the extra second was added to Coordinated Universal Time (UTC).
The slight change meant several high-profile sites which rely heavily on synchronised operations were disrupted.
Computing expert Prof Alan Woodward, from the University of Surrey, explained that the leap second can knock services out-of-sync and lead to missing data.
"Across the global internet, a lot can happen in that second," he said.
Among the affected sites was social news aggregator Reddit, which had also been hit by the storm-related problems.
"You ever wish you had an extra second or two? This is not one of those times," the site tweeted.
"We are having some Java/Cassandra issues related to the leap second at 5pm PST," it had confirmed earlier.
"We're working as quickly as we can to restore service."
The site was back up again by 4:40am on Sunday morning.
Other sites knocked out included Foursquare, LinkedIn and Gawker, as well as another Reddit-like site, StumbleUpon.
Other reports suggested that airline booking systems were also disrupted.
In a blog post written last year, Google explained how it planned to avoid the leap second issue by using a tactic it called a "leap smear".
This involved incrementally adding tiny fractions of time - a couple of milliseconds - gradually over the course of a day.
"This meant that when it became time to add an extra second at midnight, our clocks had already taken this into account, by skewing the time over the course of the day," explained Christopher Pascoe, the company's site reliability engineer.