I saw a search request take far longer than the timeout value set (multiple minutes). I understand that the timeout is imprecise but that seems a bit high.
One thing I thought of -- does this timeout include time spent in queue or just time once the request starts executing?
For more context, I'm trying to figure out whether the timeout by itself is an effective way to prevent queue backup or if we need to explicitly set the the queue size too.
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.