Today our redundant network experienced two isolated server hardware failures within our NYC POP, from 11:20-12:40 GMT, and from 17:22-18:06 GMT. Customers who were connected through the affected servers automatically failed over to their secondary POP servers. Our engineering team was automatically alerted, brought the servers back online, and customers automatically reconnected back to their primary servers.
For the duration of each of these incidents, affected customers may have experienced a small increase in latency and some missing circuit health metrics in our web dashboard. Our engineers are working to investigate the root cause, but no further incidents are expected.
Posted Apr 13, 2022 - 17:40 PDT
This incident affected: Gateway Cluster: New York, NY.