MAPREDUCE-2243. Close streams propely in a finally-block to avoid leakage in CompletedJobStatusStore, TaskLog, EventWriter and TotalOrderPartitioner. Contributed by Devaraj K

git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/trunk@1152787 13f79535-47bb-0310-9956-ffa450edef68

mapreduce/CHANGES.txt

@@ -362,6 +362,10 @@ Trunk (unreleased changes)
     MAPREDUCE-2463. Job history files are not moved to done folder when job
     history location is hdfs.  (Devaraj K via szetszwo)
 
+    MAPREDUCE-2243. Close streams propely in a finally-block to avoid leakage
+    in CompletedJobStatusStore, TaskLog, EventWriter and TotalOrderPartitioner.
+    (Devaraj K via szetszwo)
+
 Release 0.22.0 - Unreleased
 
   INCOMPATIBLE CHANGES

mapreduce/src/java/org/apache/hadoop/mapred/CompletedJobStatusStore.java

@@ -32,6 +32,7 @@
 import org.apache.hadoop.security.AccessControlException;
 import org.apache.hadoop.fs.permission.FsAction;
 import org.apache.hadoop.fs.permission.FsPermission;
+import org.apache.hadoop.io.IOUtils;
 import org.apache.hadoop.util.DiskChecker.DiskErrorException;
 
 /**
@@ -172,8 +173,9 @@ public void store(JobInProgress job) {
     if (active && retainTime > 0) {
       JobID jobId = job.getStatus().getJobID();
       Path jobStatusFile = getInfoFilePath(jobId);
+      FSDataOutputStream dataOut = null;
       try {
-        FSDataOutputStream dataOut = fs.create(jobStatusFile);
+        dataOut = fs.create(jobStatusFile);
 
         job.getStatus().write(dataOut);
 
@@ -189,6 +191,7 @@ public void store(JobInProgress job) {
         }
 
         dataOut.close();
+        dataOut = null; // set dataOut to null explicitly so that close in finally will not be executed again.
       } catch (IOException ex) {
         LOG.warn("Could not store [" + jobId + "] job info : " +
                  ex.getMessage(), ex);
@@ -198,6 +201,8 @@ public void store(JobInProgress job) {
         catch (IOException ex1) {
           //ignore
         }
+      } finally {
+        IOUtils.cleanup(LOG, dataOut);
       }
     }
   }

mapreduce/src/java/org/apache/hadoop/mapred/TaskLog.java

@@ -41,6 +41,7 @@
 import org.apache.hadoop.fs.FileStatus;
 import org.apache.hadoop.fs.FileUtil;
 import org.apache.hadoop.fs.Path;
+import org.apache.hadoop.io.IOUtils;
 import org.apache.hadoop.io.SecureIOUtils;
 import org.apache.hadoop.mapreduce.JobID;
 import org.apache.hadoop.mapreduce.util.ProcessTree;
@@ -111,13 +112,17 @@ private static LogFileDetail getLogFileDetail(TaskAttemptID taskid,
     //stderr:<start-offset in the stderr file> <length>
     //syslog:<start-offset in the syslog file> <length>
     LogFileDetail l = new LogFileDetail();
-    String str = fis.readLine();
+    String str = null;
+    try {
+      str = fis.readLine();
       if (str == null) { // the file doesn't have anything
-      throw new IOException ("Index file for the log of " + taskid+" doesn't exist.");
+        throw new IOException("Index file for the log of " + taskid
+            + " doesn't exist.");
       }
-    l.location = str.substring(str.indexOf(LogFileDetail.LOCATION)+
+      l.location = str.substring(str.indexOf(LogFileDetail.LOCATION)
-        LogFileDetail.LOCATION.length());
+          + LogFileDetail.LOCATION.length());
-    //special cases are the debugout and profile.out files. They are guaranteed
+      // special cases are the debugout and profile.out files. They are
+      // guaranteed
       // to be associated with each task attempt since jvm reuse is disabled
       // when profiling/debugging is enabled
       if (filter.equals(LogName.DEBUGOUT) || filter.equals(LogName.PROFILE)) {
@@ -139,6 +144,10 @@ private static LogFileDetail getLogFileDetail(TaskAttemptID taskid,
         str = fis.readLine();
       }
       fis.close();
+      fis = null;
+    } finally {
+      IOUtils.cleanup(LOG, fis);
+    }
     return l;
   }
   
@@ -190,6 +199,7 @@ private static void writeToIndexFile(String logLocation,
     //STDOUT: <start-offset in the stdout file> <length>
     //STDERR: <start-offset in the stderr file> <length>
     //SYSLOG: <start-offset in the syslog file> <length>   
+    try{
       dos.writeBytes(LogFileDetail.LOCATION + logLocation + "\n"
           + LogName.STDOUT.toString() + ":");
       dos.writeBytes(Long.toString(prevOutLength) + " ");
@@ -205,6 +215,10 @@ private static void writeToIndexFile(String logLocation,
           .toString()).length() - prevLogLength)
           + "\n");
       dos.close();
+      dos = null;
+    } finally {
+      IOUtils.cleanup(LOG, dos);
+    }
 
     File indexFile = getIndexFile(currentTaskid, isCleanup);
     Path indexFilePath = new Path(indexFile.getAbsolutePath());

mapreduce/src/java/org/apache/hadoop/mapreduce/jobhistory/EventWriter.java

@@ -22,6 +22,7 @@
 import java.util.Iterator;
 
 import org.apache.hadoop.fs.FSDataOutputStream;
+import org.apache.hadoop.io.IOUtils;
 import org.apache.hadoop.mapreduce.Counter;
 import org.apache.hadoop.mapreduce.CounterGroup;
 import org.apache.hadoop.mapreduce.Counters;
@@ -33,6 +34,8 @@
 import org.apache.avro.specific.SpecificDatumWriter;
 import org.apache.avro.generic.GenericData;
 import org.apache.avro.util.Utf8;
+import org.apache.commons.logging.Log;
+import org.apache.commons.logging.LogFactory;
 
 /**
  * Event Writer is an utility class used to write events to the underlying
@@ -47,6 +50,7 @@ class EventWriter {
   private DatumWriter<Event> writer =
     new SpecificDatumWriter<Event>(Event.class);
   private Encoder encoder;
+  private static final Log LOG = LogFactory.getLog(EventWriter.class);
   
   EventWriter(FSDataOutputStream out) throws IOException {
     this.out = out;
@@ -72,8 +76,13 @@ void flush() throws IOException {
   }
 
   void close() throws IOException {
+    try {
       encoder.flush();
       out.close();
+      out = null;
+    } finally {
+      IOUtils.cleanup(LOG, out);
+    }
   }
 
   private static final Schema GROUPS =

mapreduce/src/java/org/apache/hadoop/mapreduce/lib/partition/TotalOrderPartitioner.java

@@ -23,6 +23,8 @@
 import java.util.ArrayList;
 import java.util.Arrays;
 
+import org.apache.commons.logging.Log;
+import org.apache.commons.logging.LogFactory;
 import org.apache.hadoop.classification.InterfaceAudience;
 import org.apache.hadoop.classification.InterfaceStability;
 import org.apache.hadoop.conf.Configurable;
@@ -30,6 +32,7 @@
 import org.apache.hadoop.fs.FileSystem;
 import org.apache.hadoop.fs.Path;
 import org.apache.hadoop.io.BinaryComparable;
+import org.apache.hadoop.io.IOUtils;
 import org.apache.hadoop.io.NullWritable;
 import org.apache.hadoop.io.SequenceFile;
 import org.apache.hadoop.io.RawComparator;
@@ -56,6 +59,7 @@ public class TotalOrderPartitioner<K extends WritableComparable<?>,V>
   public static final String NATURAL_ORDER = 
     "mapreduce.totalorderpartitioner.naturalorder";
   Configuration conf;
+  private static final Log LOG = LogFactory.getLog(TotalOrderPartitioner.class);
 
   public TotalOrderPartitioner() { }
 
@@ -298,11 +302,16 @@ private K[] readPartitions(FileSystem fs, Path p, Class<K> keyClass,
     ArrayList<K> parts = new ArrayList<K>();
     K key = ReflectionUtils.newInstance(keyClass, conf);
     NullWritable value = NullWritable.get();
+    try {
       while (reader.next(key, value)) {
         parts.add(key);
         key = ReflectionUtils.newInstance(keyClass, conf);
       }
       reader.close();
+      reader = null;
+    } finally {
+      IOUtils.cleanup(LOG, reader);
+    }
     return parts.toArray((K[])Array.newInstance(keyClass, parts.size()));
   }