Hi all,
I’m new to Lucene and I’m evaluating it in a web application which looks up strings in a huge index – the index file contains 32GB. I keep a reference to a Searcher object during the application’s lifetime, but this object has strong memory requirements and keeps memory consumption around 950MB. I did some optimization in order to share some fields in two “composed” indices, but in a web application with less than 1GB for JVM, OutOfMemoryError is generated. It seems that the searcher keeps some form of cache which is not frequently released. I’d like to know if this kind of memory leak is normal according to Lucene’s behaviour and if the only available solution is adding memory to the JVM. Thanks in advance! -- Monique Monteiro, MSc IBM OOAD / SCJP / MCTS Web Blog: http://moniquelouise.spaces.live.com/ Twitter: http://twitter.com/monilouise MSN: monique_lou...@msn.com GTalk: monique.lou...@gmail.com
