TestIndexWriterReader fails: too many open files [LUCENE-2873]

```
[junit] Testsuite: org.apache.lucene.index.TestIndexWriterReader
[junit] Testcase: testAddIndexesAndDoDeletesThreads(org.apache.lucene.index.TestIndexWriterReader): Caused an ERROR
[junit] /home/cron/branch_3x/lucene/build/test/6/test7430286492423218781tmp/_90.prx (Too many open files)
[junit] java.io.FileNotFoundException: /home/cron/branch_3x/lucene/build/test/6/test7430286492423218781tmp/_90.prx (Too many open files)
[junit] at java.io.RandomAccessFile.open(Native Method)
[junit] at java.io.RandomAccessFile.(RandomAccessFile.java:229)
[junit] at org.apache.lucene.store.SimpleFSDirectory$SimpleFSIndexInput$Descriptor.(SimpleFSDirectory.java:69)
[junit] at org.apache.lucene.store.SimpleFSDirectory$SimpleFSIndexInput.(SimpleFSDirectory.java:90)
[junit] at org.apache.lucene.store.NIOFSDirectory$NIOFSIndexInput.(NIOFSDirectory.java:91)
[junit] at org.apache.lucene.store.NIOFSDirectory.openInput(NIOFSDirectory.java:78)
[junit] at org.apache.lucene.store.FSDirectory.openInput(FSDirectory.java:353)
[junit] at org.apache.lucene.store.MockDirectoryWrapper.openInput(MockDirectoryWrapper.java:358)
[junit] at org.apache.lucene.store.Directory.openInput(Directory.java:139)
[junit] at org.apache.lucene.index.SegmentReader$CoreReaders.(SegmentReader.java:135)
[junit] at org.apache.lucene.index.SegmentReader.get(SegmentReader.java:583)
[junit] at org.apache.lucene.index.SegmentReader.get(SegmentReader.java:561)
[junit] at org.apache.lucene.index.DirectoryReader.(DirectoryReader.java:101)
[junit] at org.apache.lucene.index.ReadOnlyDirectoryReader.(ReadOnlyDirectoryReader.java:27)
[junit] at org.apache.lucene.index.DirectoryReader$1.doBody(DirectoryReader.java:78)
[junit] at org.apache.lucene.index.SegmentInfos$FindSegmentsFile.run(SegmentInfos.java:697)
[junit] at org.apache.lucene.index.DirectoryReader.open(DirectoryReader.java:72)
[junit] at org.apache.lucene.index.IndexReader.open(IndexReader.java:344)
[junit] at org.apache.lucene.index.IndexReader.open(IndexReader.java:230)
[junit] at org.apache.lucene.index.TestIndexWriterReader.testAddIndexesAndDoDeletesThreads(TestIndexWriterReader.java:381)
[junit] at org.apache.lucene.util.LuceneTestCase$LuceneTestCaseRunner.runChild(LuceneTestCase.java:1007)
[junit] at org.apache.lucene.util.LuceneTestCase$LuceneTestCaseRunner.runChild(LuceneTestCase.java:939)
[junit]
[junit]
[junit] Tests run: 18, Failures: 0, Errors: 1, Time elapsed: 13.56 sec
[junit]
[junit] ------------- Standard Error -----------------
[junit] NOTE: reproduce with: ant test -Dtestcase=TestIndexWriterReader -Dtestmethod=testAddIndexesAndDoDeletesThreads -Dtests.seed=-7781539944268912038:-6865031686554264582
[junit] NOTE: test params are: locale=ar_SD, timezone=Asia/Almaty
[junit] NOTE: all tests run in this JVM:
[junit] [TestMergeSchedulerExternal, TestCharFilter, TestISOLatin1AccentFilter, TestCharTermAttributeImpl, TestDoc, TestFieldsReader, TestFilterIndexReader, TestIndexWriterReader]
[junit] NOTE: Linux 2.6.32-24-generic x86/IBM Corporation 1.6.0 (32-bit)/cpus=1,threads=3,free=6495816,total=11920384
[junit] ------------- ---------------- ---------------
[junit] TEST org.apache.lucene.index.TestIndexWriterReader FAILED
```

---
Migrated from [LUCENE-2873](https://issues.apache.org/jira/browse/LUCENE-2873) by Robert Muir (@rmuir), updated Oct 14 2015
Environment:
```
java version "1.6.0"
Java(TM) SE Runtime Environment (build pxi3260sr9-20101125_01(SR9))
IBM J9 VM (build 2.4, JRE 1.6.0 IBM J9 2.4 Linux x86-32 jvmxi3260sr9-20101124_69295 (JIT enabled, AOT enabled)
J9VM - 20101124_069295
JIT - r9_20101028_17488ifx2
GC - 20101027_AA)
JCL - 20101119_01

```

Start by reproducing org.apache.lucene.index.TestIndexWriterReader.testAddIndexesAndDoDeletesThreads with ant test -Dtestcase=TestIndexWriterReader -Dtestmethod=testAddIndexesAndDoDeletesThreads and the recorded seed. Trace the open-input path through NIOFSDirectory, MockDirectoryWrapper, SegmentReader, and DirectoryReader; done means the test no longer errors with Too many open files.

Written by the indexing model from the issue text.