I am using ES0.90.0.Beta1 and make search in java api, while I found I
always could not get the whole search hits set.
For example, the size of search hits is 824 using prepareCount api, while I
could only acquire 822 hits using prepareSearch api with the same query.
so in my program, after I get the size of search hits using countSearch,
then I want to acquire all the results using prepareSearch.
At first, I thought it was caused by the timeout reason, but the problem
still exist even I set a timeout value for prepareSearch.
By the way, I am conducting search in an index including three indices all
with a same alias, would this affect?
Have anybody encountered this problem before.
What value did you use as from argument in your search request?
What is the total hits count for the search request?
Did the data in your indices change between your count and search request?
A better mechanism to fetch all documents is the use the search type scan
instead of doing a count and several normal search requests:
I am using ES0.90.0.Beta1 and make search in java api, while I found I
always could not get the whole search hits set.
For example, the size of search hits is 824 using prepareCount api, while
I could only acquire 822 hits using prepareSearch api with the same query.
so in my program, after I get the size of search hits using countSearch,
then I want to acquire all the results using prepareSearch.
At first, I thought it was caused by the timeout reason, but the problem
still exist even I set a timeout value for prepareSearch.
By the way, I am conducting search in an index including three indices all
with a same alias, would this affect?
Have anybody encountered this problem before.
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.