Twitter apologizes for outage, blames bug in latest release [Update]

For a brief time today, Twitter went down leaving us all without our fix of 140 character messages. The social networking service suffered an outage that lasted for a few hours earlier this afternoon. While Twitter users are used to issues where the service goes down due to excessive traffic for a few seconds or minutes, this outage today was much longer and affected both the Twitter.com website as well as the mobile app versions.

Full Twitter service seems to be restored, at least for now. The company's "Twitter Comms" page has posted up a couple of messages since then. One said, "Today's outage is due to a cascaded bug in one of our infrastructure components. We'll provide updated information soon."

The other message offered a little more info, saying, "A cascaded bug has an effect that isn’t confined to a particular software element; its effect 'cascades" into other elements as well."

Update: And you can read the full extent of the issues relating to the downtime by clicking the image below, or link below it.

Link: Twitter Comms downtime statement
Source: Twitter Comms page

Report a problem with article
Previous Story

Google accidentally leaks Android 4.1 debut

Next Story

Fleksy aims to make iOS usable for the visually impaired

15 Comments

Commenting is disabled on this article.

Totalaero said,
Twitter use Mac Servers so I guess this is all they can say.

Sure, because writing dodgy code is Apple's fault

Totalaero said,
I was reading from here http://twitter.com/jobs/positions?jvi=oHJfWfwl,Job I know it says a plus but it means they do have apple servers but it's not completely run on apple servers.

That's an IT Admin job. Huge difference between that IT Admin job and one of the sys admins. That IT Admin handles their office's technology problems, not the twitter.com service. That job also has Windows Server experience listed, but you didn't mention that.

Feinraf said,
Tweets never stopped coming in from Tweetdeck all day?

Maybe only the frontend went down and the backend stayed mostly available