Alexa muted, Twilio taps out, and Bitbucket kicks the, er, bucket amid AWS data center hiccup

Ah, the stable and reliable cloud

By Shaun Nichols in San Francisco

Posted in Cloud, 2nd March 2018 18:52 GMT

Problems at an Amazon Web Service data center in Virginia today are being blamed for outages affecting some major websites and online services today.

Amazon said its highly popular US-East region has been dropping packets all morning, US West Coast time, thanks to an unidentified connectivity problem with Direct Connect. That's the system that connects Amazon's data centers to non-Amazon server warehouses and premises. AWS said its engineers are working on the problem, but gave no estimate when service might return.

"We continue to work towards resolving the increased packet loss impacting AWS Direct Connect connectivity to the US-EAST-1 Region. Direct Connect connections from the Equinix DC1 - DC6 & DC10 - DC12, Ashburn, VA and CoreSite VA1 & VA2, Reston, VA locations are affected by this issue," Amazon said on its status page.

Here's the full text from Amazon's status page:

7:29 AM PST We are investigating increased packet loss possibly impacting some AWS Direct Connect customers in the US-EAST-1 Region.

8:03 AM PST We continue to work towards resolving the increased packet loss impacting AWS Direct Connect connectivity to the US-EAST-1 Region. Direct Connect connections from the Equinix DC1 - DC6 & DC10 - DC12, Ashburn, VA and CoreSite VA1 & VA2, Reston, VA locations are affected by this issue.

9:43 AM PST Connections in the CoreSite VA1 & VA2 (Reston, VA) location, and some connections in the Equinix DC1 - DC6 & DC10 - DC12 (Ashburn, VA) location are inactive. Inactive connections are not receiving routes advertised from Direct Connect routers. We are now working to restore service on these Direct Connect connections. The AWS VPN service is operating normally and may be an alternative for some workloads.

The cockup has knocked a number of internet services offline, including Amazon's own Alexa virtual assistant. Others reportedly knackered by the issue include bank Capital One, comms service Twilio, and developer hub BitBucket, which has happily confirmed that the Bezos Bunch's wonky data center is at fault.

At least users are taking the outage in stride:

No word yet on when service will be restored. We've asked Amazon for comment on any possible cause for the issue, particularly if it was at all connected to the recent wave of memcache DDoS attacks. ®

Sign up to our NewsletterGet IT in your inbox daily

7 Comments

More from The Register

Amazon and others sniffing around Slack

Enjoying a free version of the biz collab software? Maybe not for much longer...

KVM? Us? Amazon erases new hypervisor from AWS EC2 FAQ

We've fro-Xen page to preserve evidence of NVMe servers and Xen's stay of execution

Slack cuts ties to IRC and XMPP, cos they don't speak Emoji

Gateways to close on May 15th, leaving you almost ten whole weeks to rebuild integrations

Corporate IMer Slack pockets $250m from SoftBank and pals – report

Reports throw cold water on Amazon buyout rumours

Hipsterverse horror as Slack takes Halloween hiatus

Two-hour outage sees users forced to speak to each other

Chap tames Slack by piping it into Emacs

And then filtering out all the stuff he doesn't need to read

AWS baits cloud hooks with DeepLens machine learning camera

Shipping in June for diehard devs with a lust for IoT kit

Amazon: Intel Meltdown patch will slow down your AWS EC2 server

Sysadmins notice performance dip amid security fix rollout. Not everyone hit hard. YMMV etc

AWS sells local Chinese infrastructure to local partner Sinnet

Bezos' cut price bit barns sell to comply with local laws

Slack apologises to Europe for TITSUP* services

Productivity melts, economy dives... actually, neither is true