10-10-2016 06:43 PM
I'm using Solr in SolrCloud mode (2 nodes out of 18 total) on CDH 5.7 Enterprise cluster for some testing. I also have Hue 3.9 configured to enable Cloudera Search to use this SolrCloud on the same cluster. When using Hue Web app GUI Indexer to index some test *.CSV files (200K rows or so) the resulting Solr collection/index that is created with single shard only allocated to a random Solr node out of SolrCloud cluster. There no option to specify number of shards to use for new index/collection at creation time in Hue Search App. Is there a way to specify/force default sharding (say to match number of Solr nodes in current SolrCloud configuration) for indexes/collections created via Hue Search Web application?