Between 00:30 UTC and 16:18 UTC some log messages were not properly delivered because our internal logging service hard crashed and had to be manually restarted.
We are digging through our log files but have yet to find the root cause.