News

The content has been bookmarked!

There was an error bookmarking this content! Please retry.

Twitter Experiences Site Instability Following Google, Microsoft Outages

Posted by Michael Floyd on May 14, 2011

Sections: Operations & Infrastructure,; Architecture & Design
Topics: Twitter API ,; OAuth ,; Public APIs ,; Authorization ,; Operations ,; API ,; Identity Management ,; Cloud Computing ,; Security ,; Infrastructure ,; Programming

Twitter is the latest to experience downtime when yesterday the company issued a status update indicating instability within the site. This was the second such report from Twitter this week and follows on the heels of outages experienced this week by Google’s Blogger service and Microsoft’s Business Productivity Online Suite (BPOS).

Mid morning yesterday Twitter issued a status report stating that:

We are currently experiencing site stability issues. There may be intermittent issues loading twitter.com and Twitter clients. We’re working to fix it as soon as possible.

Reports from Twitter’s public status page, which is monitored by Watchmouse, indicated there were also problems accessing Twitter API’s over HTTP related to OAuth and/or the statuses/home_timeline method. Calling statuses/home_timeline with a GET request requires the requesting application to authenticate the user, then returns the 20 most recent status updates including retweets.

A similar message was displayed on the Twitter status page on Tuesday, with Watchmouse reporting similar OAuth API availability issues. Anectdotally, InfoQ experienced intermittent problems on Friday when accessing Twitter.

Earlier in the week Google apologized for problems associated with Blogger. In a blog post Eddie Kessler, Technical Lead for Blogger, explained that:

...during scheduled maintenance work Wednesday night, we experienced some data corruption that impacted Blogger’s behavior.

Google rolled back the blogger service to a pre-maintenance state and placed the service into read-only mode while the company worked through the problem, making it impossible to post to Blogger.

Microsoft also offered up an explanation for outages related to BPOS, pointing to both an Exchange problem handling malformed email, and an unrelated DNS server problem. Both the Microsoft and Google outages were reported on by InfoWeek.

Last month Cloud Foundry experienced storage failure problems on the heels of the widely publicized Amazon EC2 outage.