One of the disks on Web126 is failing and driving up system load, causing degraded performance.

We’re scheduling an emergency disk swap and will update this post when we have more information.

2013-04-29 19:11 UTC: The failed disk on Web126 has been replaced and server load has returned to normal levels.