/// <summary>
/// Make sure that clients will receive StandbyExceptions even when a
/// checkpoint is in progress on the SBN, and therefore the StandbyCheckpointer
/// thread will have FSNS lock.
/// </summary>
/// <remarks>
/// Make sure that clients will receive StandbyExceptions even when a
/// checkpoint is in progress on the SBN, and therefore the StandbyCheckpointer
/// thread will have FSNS lock. Regression test for HDFS-4591.
/// </remarks>
/// <exception cref="System.Exception"/>
public virtual void TestStandbyExceptionThrownDuringCheckpoint()
{
// Set it up so that we know when the SBN checkpoint starts and ends.
FSImage spyImage1 = NameNodeAdapter.SpyOnFsImage(nn1);
GenericTestUtils.DelayAnswer answerer = new GenericTestUtils.DelayAnswer(Log);
Org.Mockito.Mockito.DoAnswer(answerer).When(spyImage1).SaveNamespace(Org.Mockito.Mockito
.Any <FSNamesystem>(), Org.Mockito.Mockito.Eq(NNStorage.NameNodeFile.Image), Org.Mockito.Mockito
.Any <Canceler>());
// Perform some edits and wait for a checkpoint to start on the SBN.
DoEdits(0, 1000);
nn0.GetRpcServer().RollEditLog();
answerer.WaitForCall();
NUnit.Framework.Assert.IsTrue("SBN is not performing checkpoint but it should be."
, answerer.GetFireCount() == 1 && answerer.GetResultCount() == 0);
// Make sure that the lock has actually been taken by the checkpointing
// thread.
ThreadUtil.SleepAtLeastIgnoreInterrupts(1000);
try
{
// Perform an RPC to the SBN and make sure it throws a StandbyException.
nn1.GetRpcServer().GetFileInfo("/");
NUnit.Framework.Assert.Fail("Should have thrown StandbyException, but instead succeeded."
);
}
catch (StandbyException se)
{
GenericTestUtils.AssertExceptionContains("is not supported", se);
}
// Make sure new incremental block reports are processed during
// checkpointing on the SBN.
NUnit.Framework.Assert.AreEqual(0, cluster.GetNamesystem(1).GetPendingDataNodeMessageCount
());
DoCreate();
Sharpen.Thread.Sleep(1000);
NUnit.Framework.Assert.IsTrue(cluster.GetNamesystem(1).GetPendingDataNodeMessageCount
() > 0);
// Make sure that the checkpoint is still going on, implying that the client
// RPC to the SBN happened during the checkpoint.
NUnit.Framework.Assert.IsTrue("SBN should have still been checkpointing.", answerer
.GetFireCount() == 1 && answerer.GetResultCount() == 0);
answerer.Proceed();
answerer.WaitForResult();
NUnit.Framework.Assert.IsTrue("SBN should have finished checkpointing.", answerer
.GetFireCount() == 1 && answerer.GetResultCount() == 1);
}
/// <exception cref="System.Exception"/>
public virtual void TestReadsAllowedDuringCheckpoint()
{
// Set it up so that we know when the SBN checkpoint starts and ends.
FSImage spyImage1 = NameNodeAdapter.SpyOnFsImage(nn1);
GenericTestUtils.DelayAnswer answerer = new GenericTestUtils.DelayAnswer(Log);
Org.Mockito.Mockito.DoAnswer(answerer).When(spyImage1).SaveNamespace(Org.Mockito.Mockito
.Any <FSNamesystem>(), Org.Mockito.Mockito.Any <NNStorage.NameNodeFile>(), Org.Mockito.Mockito
.Any <Canceler>());
// Perform some edits and wait for a checkpoint to start on the SBN.
DoEdits(0, 1000);
nn0.GetRpcServer().RollEditLog();
answerer.WaitForCall();
NUnit.Framework.Assert.IsTrue("SBN is not performing checkpoint but it should be."
, answerer.GetFireCount() == 1 && answerer.GetResultCount() == 0);
// Make sure that the lock has actually been taken by the checkpointing
// thread.
ThreadUtil.SleepAtLeastIgnoreInterrupts(1000);
// Perform an RPC that needs to take the write lock.
Sharpen.Thread t = new _Thread_404(this);
t.Start();
// Make sure that our thread is waiting for the lock.
ThreadUtil.SleepAtLeastIgnoreInterrupts(1000);
NUnit.Framework.Assert.IsFalse(nn1.GetNamesystem().GetFsLockForTests().HasQueuedThreads
());
NUnit.Framework.Assert.IsFalse(nn1.GetNamesystem().GetFsLockForTests().IsWriteLocked
());
NUnit.Framework.Assert.IsTrue(nn1.GetNamesystem().GetCpLockForTests().HasQueuedThreads
());
// Get /jmx of the standby NN web UI, which will cause the FSNS read lock to
// be taken.
string pageContents = DFSTestUtil.UrlGet(new Uri("http://" + nn1.GetHttpAddress()
.GetHostName() + ":" + nn1.GetHttpAddress().Port + "/jmx"));
NUnit.Framework.Assert.IsTrue(pageContents.Contains("NumLiveDataNodes"));
// Make sure that the checkpoint is still going on, implying that the client
// RPC to the SBN happened during the checkpoint.
NUnit.Framework.Assert.IsTrue("SBN should have still been checkpointing.", answerer
.GetFireCount() == 1 && answerer.GetResultCount() == 0);
answerer.Proceed();
answerer.WaitForResult();
NUnit.Framework.Assert.IsTrue("SBN should have finished checkpointing.", answerer
.GetFireCount() == 1 && answerer.GetResultCount() == 1);
t.Join();
}
/// <summary>
/// Test for the case when the SBN is configured to checkpoint based
/// on a time period, but no transactions are happening on the
/// active.
/// </summary>
/// <remarks>
/// Test for the case when the SBN is configured to checkpoint based
/// on a time period, but no transactions are happening on the
/// active. Thus, it would want to save a second checkpoint at the
/// same txid, which is a no-op. This test makes sure this doesn't
/// cause any problem.
/// </remarks>
/// <exception cref="System.Exception"/>
public virtual void TestCheckpointWhenNoNewTransactionsHappened()
{
// Checkpoint as fast as we can, in a tight loop.
cluster.GetConfiguration(1).SetInt(DFSConfigKeys.DfsNamenodeCheckpointPeriodKey,
0);
cluster.RestartNameNode(1);
nn1 = cluster.GetNameNode(1);
FSImage spyImage1 = NameNodeAdapter.SpyOnFsImage(nn1);
// We shouldn't save any checkpoints at txid=0
Sharpen.Thread.Sleep(1000);
Org.Mockito.Mockito.Verify(spyImage1, Org.Mockito.Mockito.Never()).SaveNamespace(
(FSNamesystem)Org.Mockito.Mockito.AnyObject());
// Roll the primary and wait for the standby to catch up
HATestUtil.WaitForStandbyToCatchUp(nn0, nn1);
Sharpen.Thread.Sleep(2000);
// We should make exactly one checkpoint at this new txid.
Org.Mockito.Mockito.Verify(spyImage1, Org.Mockito.Mockito.Times(1)).SaveNamespace
((FSNamesystem)Org.Mockito.Mockito.AnyObject(), Org.Mockito.Mockito.Eq(NNStorage.NameNodeFile
.Image), (Canceler)Org.Mockito.Mockito.AnyObject());
}
