generally speaking, how does Elasticsearch handles searching "priority" while indexing ?
I noticed that when someone performs a query with a timerange of multiple weeks, indexing almost halts while the query is running.
I find it odd that the indexing process can be disrupted by read queries.
As far as I know there is no defined or configurable priority, so reads and writes to the cluster will affect each other and compete for the same resources. A lot of use cases are more focussed on search than ingestion, so would likely make the opposite argument that ingest should be allowed to affect queries. There is therefore no defined priority that would fit everyone as it would depend a lot on the use case.
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.