Hmm, that is odd.
This is actually expected, if you are using ConcurrentMergeScheduler,
because close() by default waits for all running merges to complete.
But you seem to be using SerialMergeScheduler.
Were there any exceptions hit before you call close()?
Mike
Stu Hood <[EMAIL PROTECTED]> wrote:
> Hello,
>
> I'm having an issue with the IndexWriter in Lucene 2.3.1. Specifically, the
> IndexWriter.close() method is non-deterministically hanging with the
> following stack:
>
> """
> Thread 23044: (state = BLOCKED)
> - java.lang.Object.wait(long) @bci=0 (Compiled frame; information may be
> imprecise)
> - java.lang.Object.wait() @bci=2, line=485 (Interpreted frame)
> - org.apache.lucene.index.IndexWriter.finishMerges(boolean) @bci=306,
> line=2063 (Interpreted frame)
> - org.apache.lucene.index.IndexWriter.closeInternal(boolean) @bci=35,
> line=1208 (Interpreted frame)
> - org.apache.lucene.index.IndexWriter.close(boolean) @bci=41, line=1178
> (Interpreted frame)
> - org.apache.lucene.index.IndexWriter.close() @bci=2, line=1153
> (Interpreted frame)
> - us.webmail.solr.merge.MergeTask.execute() @bci=234, line=237 (Interpreted
> frame)
> - us.webmail.solr.merge.MergeAgent.run(java.lang.String[]) @bci=455,
> line=521 (Interpreted frame)
> -
> org.apache.hadoop.util.ToolRunner.run(org.apache.hadoop.conf.Configuration,
> org.apache.hadoop.util.Tool, java.lang.String[]) @bci=38, line=65
> (Interpreted frame)
> -
> org.apache.hadoop.util.ToolBase.doMain(org.apache.hadoop.conf.Configuration,
> java.lang.String[]) @bci=3, line=54 (Interpreted frame)
> - us.webmail.solr.merge.MergeAgent.main(java.lang.String[]) @bci=17,
> line=572 (Interpreted frame)
> - sun.reflect.NativeMethodAccessorImpl.invoke0(java.lang.reflect.Method,
> java.lang.Object, java.lang.Object[]) @bci=0 (Interpreted frame)
> - sun.reflect.NativeMethodAccessorImpl.invoke(java.lang.Object,
> java.lang.Object[]) @bci=87, line=39 (Interpreted frame)
> - sun.reflect.DelegatingMethodAccessorImpl.invoke(java.lang.Object,
> java.lang.Object[]) @bci=6, line=25 (Interpreted frame)
> - java.lang.reflect.Method.invoke(java.lang.Object, java.lang.Object[])
> @bci=161, line=597 (Interpreted frame)
> - org.apache.hadoop.util.RunJar.main(java.lang.String[]) @bci=538, line=155
> (Interpreted frame)
> """
>
>
> My IndexWriter opening code looks like this:
> """
> // open the destination index
> this.targetDirectory = null;
> try{
> this.targetDirectory =
> FSDirectory.getDirectory(getDD(this.targetInstance),
> new NativeFSLockFactory(getDD(this.targetInstance)));
> this.targetIndex = new IndexWriter(this.targetDirectory, null, false);
> this.targetIndex.setMergeFactor(MERGE_FACTOR);
> this.targetIndex.setUseCompoundFile(false);
> // this.targetIndex.setInfoStream(System.out);
>
> this.targetIndex.setMergeScheduler(new SerialMergeScheduler());
>
>
> } catch (Exception e){
> if(this.targetDirectory != null){
> this.targetDirectory.close();
> }
> throw new IOException("Could not open target instance: " + e.toString());
> }
> """
>
> After adding a bunch of indexes with IndexWriter.addIndexes(Directory[]), I
> run
> """
> System.out.println("Closing the target index...");
> this.targetIndex.close();
> System.out.println("...done.\nClosing the target directory...");
> this.targetDirectory.close();
> System.out.println("...done.");
> """
> ... and the output clearly shows that we get stuck in close.
>
> -------------------------------------------------------------
>
> Is this a known bug? Does anyone know of a workaround?
>
> Stu Hood
> Architecture Software Developer
> Mailtrust, a Division of Rackspace
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
