We use Redis between Filebeat and Logstash. We set tags in our log prospectors and send those to individual Redis lists based on the tag and to multiple Redis instances. We then set-up Logtash to read from the multiple Redis servers.
We have ten Logstash instances and 10 Redis servers. We set-up each Logstash server to read from each Redis server. Therefore if we lose one Logstash instance, the other nine can take the strain. Effectively it is a poor man's clustered Logstash.
I am not sure whether this helps but it may give you another angle to look at.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.