The log message indicates that the search thread pool is at capacity of 1000, and so subsequent search requests are being rejected. Are you monitoring the cluster to understand the search load?
How many queries are you typically servicing concurrently? How many shards do these queries typically address? How many indices and shards do you have per node?
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.