All systems have returned to normal for all components. As of the last update, all message sending was caught up and back to real-time for all customers.
At this time, data processing has caught up on the backlog for affected customers and is back to real-time. No device data was lost.
Posted Nov 11, 2017 - 15:16 EST
Update
Message sending for affected customers has fully caught back up to real-time and is no longer delayed. All queued messages have been sent. Data processing for affected customers is being released from queue and processing now.
Posted Nov 11, 2017 - 14:46 EST
Update
A subset of customers may also be experiencing delayed message sending.
Posted Nov 11, 2017 - 10:07 EST
Identified
Our DBA team has identified the source of the issue and are working toward resolution.
Posted Nov 11, 2017 - 09:29 EST
Investigating
We are seeing elevated error rates with SDK data collection for the US 01 cluster. Data processing is delayed by a few minutes for some customers on US 01 as well.