A critical component of Laboperator was unexpectedly restarted on midnight, which resulted in some Laboperator worker processes not being able to process incoming messages, including incoming data points from devices. Restarting said processes force it to reestablish said connection, and the workers was able to caught up with the queued messages.
Given that the issue was affecting only the workers processes, we did not identify any resulting data gaps in the datapoints collected from devices.