Browse Source

HDFS-7355. TestDataNodeVolumeFailure#testUnderReplicationAfterVolFailure fails on Windows, because we cannot deny access to the file owner. Contributed by Chris Nauroth.

Haohui Mai 10 years ago
parent
commit
99d710348a

+ 4 - 0
hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt

@@ -1003,6 +1003,10 @@ Release 2.6.0 - UNRELEASED
     HDFS-7334. Fix periodic failures of TestCheckpoint
     HDFS-7334. Fix periodic failures of TestCheckpoint
     #testTooManyEditReplayFailures. (Charles Lamb via wheat9)
     #testTooManyEditReplayFailures. (Charles Lamb via wheat9)
 
 
+    HDFS-7355. TestDataNodeVolumeFailure#testUnderReplicationAfterVolFailure
+    fails on Windows, because we cannot deny access to the file owner.
+    (Chris Nauroth via wheat9)
+
     BREAKDOWN OF HDFS-6134 AND HADOOP-10150 SUBTASKS AND RELATED JIRAS
     BREAKDOWN OF HDFS-6134 AND HADOOP-10150 SUBTASKS AND RELATED JIRAS
   
   
       HDFS-6387. HDFS CLI admin tool for creating & deleting an
       HDFS-6387. HDFS CLI admin tool for creating & deleting an

+ 5 - 0
hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/datanode/TestDataNodeVolumeFailure.java

@@ -205,6 +205,11 @@ public class TestDataNodeVolumeFailure {
    */
    */
   @Test
   @Test
   public void testUnderReplicationAfterVolFailure() throws Exception {
   public void testUnderReplicationAfterVolFailure() throws Exception {
+    // This test relies on denying access to data volumes to simulate data volume
+    // failure.  This doesn't work on Windows, because an owner of an object
+    // always has the ability to read and change permissions on the object.
+    assumeTrue(!Path.WINDOWS);
+
     // Bring up one more datanode
     // Bring up one more datanode
     cluster.startDataNodes(conf, 1, true, null, null);
     cluster.startDataNodes(conf, 1, true, null, null);
     cluster.waitActive();
     cluster.waitActive();