Seems that it means the indexing operations are occurring on this node, but that's too huge for me cause the real indexing speed value is very small no more than 500. Any idea?
SomeOne ask the same question five years ago ,but no answer,I do not know why...
Es version is 2.3.2.
Another ES cluster(6.0.0)which real indexing QPS value up to 400K per-second,I using the same API ,return result is about 100.
Very ridiculous,can u tell me why?
Ok, there might have been a bugfix within the last three and a half years, so I will not investigate that further.
The second part can be explained somewhat simpler. This is a very small point in time view, where the statistics are taken in the point in time where indexing happens, so there are taken in a certain millisecond of a second (or even micro- or nanosecond) and during that point in time, 100 documents were indexed.
Sounds reasonable.I wonder where the real time monitoring metric data came from such as " Search Rate (/s)"," Indexing Rate (/s)" in kibana webUI. Official said that they were from API like "_stats,node/_stats" .Can u tell me about that ,thks very much.
you are right with your assumption that this should be the number of current indexing operations, and that indeed sounds pretty high.
One source of confusion that I've seen here in the past is that people assume that one indexing operation is the same as one document being ingested, which is not the case. This counter is incremented anytime an index operation happens, which can include many types of internal operations.
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.