The message says that this could be due to large docs or slow I/O or too many docs, but seems odd that it is just happening on this one node and the other looks fine?
this is impossible to say with just looking at a single number. There could be many reasons for that, like the node holding more data, the node processing more queries, the node indexing more heavy while answering queries. You could have slower disk in that one. etc
You have not mentioned if the other nodes are hit with the exact same load. Are you maybe always hitting this node for your search queries. Have you checked the nodes stats/nodes info output for any anomalies compared with the other nodes?
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.