HDFS-2994. If lease soft limit is recovered successfully the append can fail. Contributed by Tao Luo.

git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/trunk@1514500 13f79535-47bb-0310-9956-ffa450edef68

hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt

@@ -342,6 +342,9 @@ Release 2.1.1-beta - UNRELEASED
     HDFS-5099. Namenode#copyEditLogSegmentsToSharedDir should close
     EditLogInputStreams upon finishing. (Chuan Liu via cnauroth)
 
+    HDFS-2994. If lease soft limit is recovered successfully
+    the append can fail. (Tao Luo via shv)
+
 Release 2.1.0-beta - 2013-08-22
 
   INCOMPATIBLE CHANGES

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSNamesystem.java

@@ -2153,10 +2153,15 @@ public class FSNamesystem implements Namesystem, FSClusterStats,
         throw new FileNotFoundException("failed to append to non-existent file "
           + src + " on client " + clientMachine);
       }
-      final INodeFile myFile = INodeFile.valueOf(inode, src, true);
+      INodeFile myFile = INodeFile.valueOf(inode, src, true);
       // Opening an existing file for write - may need to recover lease.
       recoverLeaseInternal(myFile, src, holder, clientMachine, false);
-
+      
+      // recoverLeaseInternal may create a new InodeFile via 
+      // finalizeINodeFileUnderConstruction so we need to refresh 
+      // the referenced file.  
+      myFile = INodeFile.valueOf(dir.getINode(src), src, true);
+      
       final DatanodeDescriptor clientNode = 
           blockManager.getDatanodeManager().getDatanodeByHost(clientMachine);
       return prepareFileForWrite(src, myFile, holder, clientMachine, clientNode,

hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestFileAppend.java

@@ -17,6 +17,7 @@
  */
 package org.apache.hadoop.hdfs;
 
+import static org.junit.Assert.assertEquals;
 import static org.junit.Assert.assertTrue;
 
 import java.io.File;
@@ -337,4 +338,47 @@ public class TestFileAppend{
       cluster.shutdown();
     }
   }
+  
+  /** Tests appending after soft-limit expires. */
+  @Test
+  public void testAppendAfterSoftLimit() 
+      throws IOException, InterruptedException {
+    Configuration conf = new HdfsConfiguration();
+    conf.setInt(DFSConfigKeys.DFS_REPLICATION_KEY, 1);
+    conf.setBoolean(DFSConfigKeys.DFS_SUPPORT_APPEND_KEY, true);
+    //Set small soft-limit for lease
+    final long softLimit = 1L;
+    final long hardLimit = 9999999L;
+
+    MiniDFSCluster cluster = new MiniDFSCluster.Builder(conf).numDataNodes(1)
+        .build();
+    cluster.setLeasePeriod(softLimit, hardLimit);
+    cluster.waitActive();
+
+    FileSystem fs = cluster.getFileSystem();
+    FileSystem fs2 = new DistributedFileSystem();
+    fs2.initialize(fs.getUri(), conf);
+
+    final Path testPath = new Path("/testAppendAfterSoftLimit");
+    final byte[] fileContents = AppendTestUtil.initBuffer(32);
+
+    // create a new file without closing
+    FSDataOutputStream out = fs.create(testPath);
+    out.write(fileContents);
+
+    //Wait for > soft-limit
+    Thread.sleep(250);
+
+    try {
+      FSDataOutputStream appendStream2 = fs2.append(testPath);
+      appendStream2.write(fileContents);
+      appendStream2.close();
+      assertEquals(fileContents.length, fs.getFileStatus(testPath).getLen());
+    } finally {
+      fs.close();
+      fs2.close();
+      cluster.shutdown();
+    }
+  }
+
 }