Given issues such as Kafka input slow stop consuming , its clear that some people can see silent throughput failures that may be related to sinks of logstash.
What happens when a logstash sink isn't capable of consuming events fast enough, and how can we forsee such an event happening in the logs before its 'too late' ?