Currently I am dealing with a small dataset (a few thousands). Instead of updating the index(file based), I am creating a new file index with a another name when data changes. And whenever I rebuild the index in the new file, I make the Indexsearcher to point to this new file instead of the old index file. Now in a multiuser environment, how would this behave if one user is searching something and at the same time, a new index is built and the searcher now refers to this new index ?
Also is this the right way to implement if I don't want to use incremental update. -- View this message in context: http://www.nabble.com/Lucene-Concurrency-Concern-tp25654695p25654695.html Sent from the Lucene - General mailing list archive at Nabble.com.
