The events from each input are processed separately by logstash. You might be able to do it by using an elasticsearch input for one cluster, then using an elasticsearch filter to look up the same document in the other cluster.
Sound great, but can you please maybe share some config example for that, let's suppose _id from the first cluster is the field I want to search in second cluster in order to find the doc for comparing it.
@Badger could you please provide some an example of how to do it?
And does it mean that each document will be processed separately (checked/filtered)? If yes, it sounds like it might take a lot of time to check hundred of thousands doc and many pings to target cluster...
No, I cannot provide an example because I do not run elasticsearch. An API call to ES is going to be expensive compared to many other pipeline operations.
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.