Problem:
The outage today on the sync site resulted from an
unanticipated failure mode. The primary
sync web server had an issue connecting to some of its application files
resulting in application errors. Our
load balancing system did not detect the site as being down so it did not
redirect traffic to an alternate server.
Resolution:
Internal checks were added so that if this type of error
occurs in the future, the site will be taken offline to allow the traffic to be
redirected properly.
