According to my alerts Hubski was down for at about six and a half hours this morning. This was due to a power outage in our datacenter which took our servers offline for a while. This isn't the first time we've had an issue with our servers in this datacenter being taken offline, but it has been relatively rare. My hopes for the future are to finish our rewriting of hubski to allow for a more distributed architecture which won't be affected by downtime in any one datacenter; boo single points of failure.