[HDFS-11025] TestDiskspaceQuotaUpdate fails in trunk due to Bind exception

• Type: Bug
• Status: Resolved
• Priority: Minor
• Resolution: Fixed
• Fix Version/s: 2.8.0, 3.0.0-alpha2
• Hadoop Flags: Reviewed
• Assignee: Yiqun Lin
• Reporter: Yiqun Lin
• Created: 18/Oct/16 13:05
• Updated: 21/Oct/16 00:26
• Resolved: 20/Oct/16 05:21

Description

The test TestDiskspaceQuotaUpdate fails sometimes after HDFS-10843, the link addresse: https://builds.apache.org/job/PreCommit-HDFS-Build/17200/testReport/. The stack infos:

 
java.net.BindException: Problem binding to [localhost:49195] java.net.BindException: Address already in use; For more details see:  http://wiki.apache.org/hadoop/BindException

I found the bind exception was happened in new test method TestDiskspaceQuotaUpdate.testQuotaIssuesWhileCommitting. The related codes:

  public void testQuotaIssuesWhileCommitting() throws Exception {
    ...
    try {
      for (int i = REPLICATION - 1; i > 0; i--) {
        dnprops.add(cluster.stopDataNode(i));
      }
      ...
    } finally {
      for (MiniDFSCluster.DataNodeProperties dnprop : dnprops) {
        cluster.restartDataNode(dnprop, true);
      }
      cluster.waitActive();
    }
  }

I think we can make a simple fix in cluster.restartDataNode(dnprop, true);. The tests in TestDiskspaceQuotaUpdate just care about that if the cluster is up and running. So I think this change will not influence the current logic,