Amazon said Friday morning that it expected that the majority of sites affected by its unexpected cloud services outage would be back up by the end of Friday, with some exceptions. (See PCMag’s analysis on why the Amazon cloud outage matters.)
Amaazon’s published round-the-clock updates overnight, advising customers that the company had brought “all hands on deck” to solve the problem. At issue, Amazon said, was a single “availability zone” in the Eastern U.S. that with “stuck” volumes of data.
Best Microsoft MCTS Training – Microsoft MCITP Training at Certkingdom.com
“We continue to see progress in recovering volumes, and have heard many additional customers confirm that they’re recovering,” Amazon wrote Friday morning. “Our current estimate is that the majority of volumes will be recovered over the next 5 to 6 hours. As we mentioned in our last post, a smaller number of volumes will require a more time consuming process to recover, and we anticipate that those will take longer to recover. We will continue to keep everyone updated as we have additional information.”
It appeared that the sites which had been affected by the Thursday outage to Amazon’s EC2 (Elastic Compute Cloud) and Amazon Web Services were functional, but that data for the affected period was in the process of being restored. That means that visitors to those sites would be able to use them, but not necessarily access comments, access rights, documents or other data affected by the outage.
Quora and Reddit, two sites that were affected by the outage, were operational at press time, but partly so. “We are slowly getting our capacity back, and as such users are being randomly granted access back to the site,” Reddit posted to the top of the site. “Please check back soon, as you may be able to log in shortly. Thanks!”
Charlie Cheever, one of the founders of Quora, also explained why some of the site’s data was still missing. “The Amazon EBS volumes where the data for Quora is stored are still not available,” he wrote on the site. “To get the site back online, we brought up a new database based on the most recent database backup we had available which was from midnight on Tuesday night. So, any writes to the database during the time between the backup and the outage (most activity on the site on Wednesday) are missing right now.
“When the volume is restored, we’ll try to merge the data from Wednesday back into the current version of the site,” Cheever added. “There will likely be some conflicts, but we think there will be graceful ways to resolve most of those.”
Foursquare, which had also been taken down by Amazon’s cloud problems, also reported that it had restored full access to the site at 1:40 AM EDT on Friday morning, taking the site down for ten minutes just to ensure there were no problems.
Hootsuite, which also had been affected by the outage, said that service had been restored at 9:25 PM on Thursday, although not all profiles were available.
At 6:18 AM PT, Amazon reported that it had begun to see “more meaningful progress” in restoring its volumes, and at 8:49 AM PT customers began telling Amazon that they were coming back online.