My understanding is that when a batch of events exits the pipeline it will get passed to multi_receive, which immediately indexes them. It calls the mixin, which calls submit, which calls safe_bulk, which call the bulk method of the client. That is the code which does size based batching, in case a single batch from the pipeline is larger than 20 MB.
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.