One of our DSL aggregation routers handling customers in California’s LATA 1 (largely the San Francisco area) started having problems at approximately 10:41pm tonight. Customers experienced a complete service outage until 10:47pm, at which point we were able to restore partial service. We’re presently working to move the affected customer’s service to alternate equipment, which will involve 10-15 minutes worth of additional downtime in the near future. At present, customers on this particular piece of aggregation gear are experiencing ~7% packet loss. We’ll update this entry as our work progresses.
-Nathan, Matt and Jared
The problem turned out to be a failing ATM port inside of our network. We’ve migrated all traffic to a hot-spare port and service has returned to normal.