@Michael Young
One last question. We have a Prod SOLR cluster using HDFS as file system. Assume following two scenarios:
1. SOLR is also running on DR. When we replicate data to DR using Snapshot/Disctcp combo, how does DR SOLR know which data belong to which index? I am guessing it doesn't. So in that case, how do we manage that?
2. SOLR is not running on DR. We replicate the data to DR. Some issue occurs in production and now we need to restore data back to Prod. Can we restore only some indexes? If yes, how is it possible since DR doesn't have any SOLR and for DR its simply some HDFS data.