How I can set what fields to use in '_id' generation?
I use logstash to fetch log files from different places. Sometimes those files are identical. So I want to prevent elasticsearch from adding dublicated log entries. As I understand, if system discover the same id, entry will not be added (or replaced - it does not matter). The simplest way to do this - exclude from '_id' generation path field. So how I can set what fields to use to generate '_id'?