Recent slowness and outages
One of our backend database servers has recently been experiencing corruption. This corruption has caused us to failover to our secondary system, which isn't as powerful. The switch, while providing stable performance, proved to be an issue because it caused a seemingly never ending string of slow requests and timeouts for those affected users.
Yesterday morning (around 9AM PST) we swapped back to our primary system after a restore and those users were once again online. We are making changes to that back-end server pair to provide better redundancy and a more stable environment.