According to official document said, if we unfortunately still use spin disk, we should set this value to 1.
I used to be set this in elasticsearch.yml but not luck in v5.4
I know this setting has been move to index level, what I confused is, if this is per index setting? What happened if I create two indices and set this value to different in each indices?
So merge scheduler is per shard? since each shard is actually a Lucene writer?
If a node with 2 indices, index A set max_thread_count=1, index B set max_thread_count=2. We will have 3 threads running when doing merge right?
I thought this is node setting.
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.