Until https://github.com/elastic/logstash/issues/2611 is resolved, i would like to use the metrics filter to push to ES the amount of syslog / lumberjack events processed by each of the logstash instances.
Does anyone know how this impacts the performance of the pipeline? I assume it doesn't slow down any event from moving between input > filter > output. I assume logstash just has a counter in memory and flushes that counter based on the configuration.