We're looking into why you saw an error, but the event was caused by SQL coming online in New York.
We're using SQL 2012 Always On Availability groups as part of our high availability architecture. How this works is the primary database server for that group (these are currently moved to Oregon due to Sandy) streams changes to the replicas.
Due to a lack of power in New York (status updates on our NY datacenter can be found here, here and here), we are keeping our equipment there off as much as we can. However, the replicas for SQL being down mean the backlog of changes needed to sync to them builds up in Oregon, at the rate of about 40GB that needs sending every 6 hours.
To relieve these logs (which we could tolerate for up to ~4 days) we spin up the New York SQL boxes to receive changes and let the logs clear out on the primary. On this latest spin up, the combination of no domain controller in New York online and the SQL boxes coming up simultaneously caused the windows clustering service to fail on NY-SQL01, and while the SQL cluster on top of this windows cluster assessed what was wrong, the databases were unavailable.
The app should have thrown a nice error page during this at least, we'll look into that.
We're by no means sitting around idle after our failover to Oregon. There are many, many things still happening, but we'll try and keep the interruptions to a minimum.