We are having issues with 7.4.1 and the data nodes getting backlogs on the management threadpool and then all cluster communication stops. Then just noticed that in the documentation https://www.elastic.co/guide/en/elasticsearch/reference/7.4/modules-threadpool.html the management thread pool is mentioned in 7.4 and gone in 7.5. Is cluster management completely different in 7.5? That would make it worth looking at.
Right now if we index more then say 20k documents/second we run into this backlog in cluster management.
Thanks for the heads-up. Note the weasel words at the top of the doc:
There are several thread pools, but the important ones include:
However the omission of the management threadpool in the 7.5 docs looks to be a mistake. These docs were added in #48543 and backported to the 7.4 branch but not, apparently the 7.5 branch. They're there again in 7.6: Thread Pool | Elasticsearch Guide [7.6] | Elastic
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.