[
https://issues.apache.org/jira/browse/LUCENE-3409?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Michael McCandless updated LUCENE-3409:
---------------------------------------
Affects Version/s: 4.0
Fix Version/s: 4.0
3.4
I found the issue: we are failing to drop pool'd readers in IW.deleteAll. I'll
commit fix shortly.
> NRT reader/writer over RAMDirectory memory leak
> -----------------------------------------------
>
> Key: LUCENE-3409
> URL: https://issues.apache.org/jira/browse/LUCENE-3409
> Project: Lucene - Java
> Issue Type: Bug
> Components: core/index
> Affects Versions: 3.0.2, 3.3, 4.0
> Reporter: tal steier
> Assignee: Michael McCandless
> Fix For: 3.4, 4.0
>
>
> with NRT reader/writer, emptying an index using:
> writer.deleteAll()
> writer.commit()
> doesn't release all allocated memory.
> for example the following code will generate a memory leak:
> /**
> * Reveals a memory leak in NRT reader/writer<br>
> *
> * The following main() does 10K cycles of:
> * <ul>
> * <li>Add 10K empty documents to index writer</li>
> * <li>commit()</li>
> * <li>open NRT reader over the writer, and immediately close it</li>
> * <li>delete all documents from the writer</li>
> * <li>commit changes to the writer</li>
> * </ul>
> *
> * Running with -Xmx256M results in an OOME after ~2600 cycles
> */
> public static void main(String[] args) throws Exception {
> RAMDirectory d = new RAMDirectory();
> IndexWriter w = new IndexWriter(d, new
> IndexWriterConfig(Version.LUCENE_33, new KeywordAnalyzer()));
> Document doc = new Document();
>
> for(int i = 0; i < 10000; i++) {
> for(int j = 0; j < 10000; ++j) {
> w.addDocument(doc);
> }
> w.commit();
> IndexReader.open(w, true).close();
> w.deleteAll();
> w.commit();
> }
>
> w.close();
> d.close();
> }
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
