We've now resolved the incident. Thanks for your patience.
We are changing our network configuration to avoid this type of downtime in the near future.
We are experiencing network outage from vSwitch that is used for internal communication with ElasticSearch.