Previously, they were not freed, just reused. As of the tips of 29x, 30x, 3x, trunk, they are now fully freed.
Mike On Sat, Jun 5, 2010 at 7:33 AM, Shay Banon <kim...@gmail.com> wrote: > I will check, just wanted to verify though, do they get freed when calling > commit, or do they still hang around? Other pools are freed when calling > commit, right? > > -shay.banon > > On Sat, Jun 5, 2010 at 12:55 PM, Michael McCandless < > luc...@mikemccandless.com> wrote: > >> They are reclaimed, but, you may be hitting one of the memory leak >> bugs recently fixed but not yet (though soon to be) released. >> >> Can you try running with the tip of whichever release you are using? >> >> Mike >> >> On Fri, Jun 4, 2010 at 10:10 PM, Shay Banon <kim...@gmail.com> wrote: >> > Hi, >> > >> > I have a question on how IndexWriter manages its memory when it comes to >> > RawPostingList. Its pretty late, so sorry if the question is obvious, but >> > the question is when does the RawPostingList array gets reclaimed or >> > shrinked? I assume they get factored in the ram balancing process, but, >> is >> > there a chance that they won't be reclaimed? Can I force them to be >> > reclaimed, I think that IndexWriter#commit does not affect that... . Here >> is >> > a link to a sample heap dump I took (I do call flush): >> > http://skitch.com/kimchy/dg3e4/eclipse-memory-analyzer. >> > >> > Thanks, >> > Shay >> > >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org >> For additional commands, e-mail: java-user-h...@lucene.apache.org >> >> > --------------------------------------------------------------------- To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org For additional commands, e-mail: java-user-h...@lucene.apache.org
