This issue has remained resolved since our last update. Further monitoring has been put in place to identify health issues with our Cassandra cluster.
Aug 13, 10:28 PDT
Removing offending Cassandra nodes and playbook triage against attached services have restored optimal latency. We're currently working on a RCA to prevent this from happening in the future.
Aug 13, 09:47 PDT
The team has removed unhealthy Cassandra nodes and is working to restore full performance to the system. Another update will follow shortly.
Aug 13, 09:30 PDT
We're currently mitigating issues with websocket latency and client API timeouts in NA. We will update in the next 30 minutes with more information.
Aug 13, 08:53 PDT