java.nio.channels.OverlappingFileLockException when set property for one vertex that indexed using lucene

[huiwenhan]

Get following error:
ERROR [com.thinkaurelius.titan.graphdb.database.StandardTitanGraph] Error while commiting index mutations for transaction [2364] on index: search
com.thinkaurelius.titan.core.TitanException: Could not execute operation due to backend exception
at com.thinkaurelius.titan.diskstorage.util.BackendOperation.execute(BackendOperation.java:44)
at com.thinkaurelius.titan.diskstorage.indexing.IndexTransaction.flushInternal(IndexTransaction.java:119)
at com.thinkaurelius.titan.diskstorage.indexing.IndexTransaction.commit(IndexTransaction.java:100)
at com.thinkaurelius.titan.diskstorage.BackendTransaction.commitIndexes(BackendTransaction.java:124)
at com.thinkaurelius.titan.graphdb.database.StandardTitanGraph.commit(StandardTitanGraph.java:639)
at com.thinkaurelius.titan.graphdb.transaction.StandardTitanTx.commit(StandardTitanTx.java:1228)
at com.sfb.wsa.cd.uptake.graph.titan.model.vertex.access.TagDao.add(TagDao.java:32)
at com.sfb.wsa.cd.uptake.profiles.jobs.ImportProfileJob.main(ImportProfileJob.java:108)
Caused by: com.thinkaurelius.titan.diskstorage.PermanentBackendException: Could not create writer
at com.thinkaurelius.titan.diskstorage.lucene.LuceneIndex.getWriter(LuceneIndex.java:111)
at com.thinkaurelius.titan.diskstorage.lucene.LuceneIndex.mutate(LuceneIndex.java:146)
at com.thinkaurelius.titan.diskstorage.indexing.IndexTransaction$1.call(IndexTransaction.java:122)
at com.thinkaurelius.titan.diskstorage.indexing.IndexTransaction$1.call(IndexTransaction.java:119)
at com.thinkaurelius.titan.diskstorage.util.BackendOperation.executeDirect(BackendOperation.java:56)
at com.thinkaurelius.titan.diskstorage.util.BackendOperation.execute(BackendOperation.java:42)
... 7 more
Caused by: org.apache.lucene.store.LockObtainFailedException: Lock obtain timed out: NativeFSLock@C:\tmp\titan\wsa\emp\lucene\tagmixedIndex\write.lock: java.nio.channels.OverlappingFileLockException
at org.apache.lucene.store.Lock.obtain(Lock.java:89)
at org.apache.lucene.index.IndexWriter.(IndexWriter.java:710)
at com.thinkaurelius.titan.diskstorage.lucene.LuceneIndex.getWriter(LuceneIndex.java:108)
... 12 more
Caused by: java.nio.channels.OverlappingFileLockException
at sun.nio.ch.SharedFileLockTable.checkList(FileLockTable.java:267)
at sun.nio.ch.SharedFileLockTable.add(FileLockTable.java:164)
at sun.nio.ch.FileChannelImpl.tryLock(FileChannelImpl.java:1078)
at java.nio.channels.FileChannel.tryLock(FileChannel.java:1165)
at org.apache.lucene.store.NativeFSLock.obtain(NativeFSLockFactory.java:148)
at org.apache.lucene.store.Lock.obtain(Lock.java:100)
... 14 more

[dalaro]

Did you call TitanFactory.open multiple times in one Java VM? If so, did you have only one graph open at a time, and did you call graph.shutdown() on all of the graphs when you were done with them?

If you can reproduce this reliably, then please show us how.

[dalaro]

Incidentally, I initially thought this could potentially be related to #761, but IndexReaders do not appear to have any bearing on the Writer locks. I might be mistaken though.

[mukil]

Google brought me here, so I just wanted to leave you this hint:

This seems to be related to using Java to initialize an existing file-db which is located on a storage formatted as MS-DOS FAT (e.g. usb stick) on Mac OS X.

See also: elastic/elasticsearch#1451

(I have this issue when trying to init my neo4j file-db with lucene)