Apache
/
hadoop
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

HDFS-11896. Non-dfsUsed will be doubled on dead node re-registration. Contributed by Brahma Reddy Battula.

This commit is contained in:
Brahma Reddy Battula 2017-07-27 12:02:57 -07:00 committed by Konstantin V Shvachko
parent 11ece0bda1
commit c4a85c694f
3 changed files with 65 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeDescriptor.java
View File

@ -331,11 +331,7 @@ public class DatanodeDescriptor extends DatanodeInfo {
}
public void resetBlocks() {
setCapacity(0);
setRemaining(0);
setBlockPoolUsed(0);
setDfsUsed(0);
setXceiverCount(0);
updateStorageStats(this.getStorageReports(), 0L, 0L, 0, 0, null);
this.invalidateBlocks.clear();
this.volumeFailures = 0;
// pendingCached, cached, and pendingUncached are protected by the
@ -384,6 +380,16 @@ public class DatanodeDescriptor extends DatanodeInfo {
public void updateHeartbeatState(StorageReport[] reports, long cacheCapacity,
long cacheUsed, int xceiverCount, int volFailures,
VolumeFailureSummary volumeFailureSummary) {
updateStorageStats(reports, cacheCapacity, cacheUsed, xceiverCount,
volFailures, volumeFailureSummary);
setLastUpdate(Time.now());
setLastUpdateMonotonic(Time.monotonicNow());
rollBlocksScheduled(getLastUpdateMonotonic());
}
private void updateStorageStats(StorageReport[] reports, long cacheCapacity,
long cacheUsed, int xceiverCount, int volFailures,
VolumeFailureSummary volumeFailureSummary) {
long totalCapacity = 0;
long totalRemaining = 0;
long totalBlockPoolUsed = 0;
@ -434,8 +440,6 @@ public class DatanodeDescriptor extends DatanodeInfo {
setCacheCapacity(cacheCapacity);
setCacheUsed(cacheUsed);
setXceiverCount(xceiverCount);
setLastUpdate(Time.now());
setLastUpdateMonotonic(Time.monotonicNow());
this.volumeFailures = volFailures;
this.volumeFailureSummary = volumeFailureSummary;
for (StorageReport report : reports) {
@ -451,7 +455,6 @@ public class DatanodeDescriptor extends DatanodeInfo {
totalDfsUsed += report.getDfsUsed();
totalNonDfsUsed += report.getNonDfsUsed();
}
rollBlocksScheduled(getLastUpdateMonotonic());
// Update total metrics for the node.
setCapacity(totalCapacity);

2
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/DataNode.java
View File

@ -1319,7 +1319,7 @@ public class DataNode extends ReconfigurableBase
// used only for testing
@VisibleForTesting
void setHeartbeatsDisabledForTests(
public void setHeartbeatsDisabledForTests(
boolean heartbeatsDisabledForTests) {
this.heartbeatsDisabledForTests = heartbeatsDisabledForTests;
}

53
hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestDeadDatanode.java
View File

@ -17,9 +17,11 @@
*/
package org.apache.hadoop.hdfs.server.namenode;
import com.google.common.base.Supplier;
import org.apache.hadoop.hdfs.server.protocol.SlowDiskReports;
import static org.junit.Assert.assertEquals;
import static org.junit.Assert.assertFalse;
import static org.junit.Assert.assertTrue;
import static org.junit.Assert.fail;
import java.io.IOException;
@ -36,6 +38,7 @@ import org.apache.hadoop.hdfs.protocol.Block;
import org.apache.hadoop.hdfs.protocol.BlockListAsLongs;
import org.apache.hadoop.hdfs.server.blockmanagement.BlockManager;
import org.apache.hadoop.hdfs.protocol.BlockType;
import org.apache.hadoop.hdfs.server.blockmanagement.DatanodeDescriptor;
import org.apache.hadoop.hdfs.server.blockmanagement.DatanodeManager;
import org.apache.hadoop.hdfs.server.blockmanagement.DatanodeStorageInfo;
import org.apache.hadoop.hdfs.server.datanode.DataNode;
@ -52,6 +55,7 @@ import org.apache.hadoop.hdfs.server.protocol.StorageBlockReport;
import org.apache.hadoop.hdfs.server.protocol.StorageReceivedDeletedBlocks;
import org.apache.hadoop.hdfs.server.protocol.StorageReport;
import org.apache.hadoop.net.Node;
import org.apache.hadoop.test.GenericTestUtils;
import org.junit.After;
import org.junit.Test;
@ -178,4 +182,53 @@ public class TestDeadDatanode {
.getDatanodeDescriptor().equals(clientNode));
}
}
@Test
public void testNonDFSUsedONDeadNodeReReg() throws Exception {
Configuration conf = new HdfsConfiguration();
conf.setInt(DFSConfigKeys.DFS_HEARTBEAT_INTERVAL_KEY, 1);
conf.setInt(DFSConfigKeys.DFS_NAMENODE_HEARTBEAT_RECHECK_INTERVAL_KEY, 1);
conf.setInt(DFSConfigKeys.DFS_NAMENODE_STALE_DATANODE_INTERVAL_KEY,
6 * 1000);
long CAPACITY = 5000L;
long[] capacities = new long[] { 4 * CAPACITY, 4 * CAPACITY };
try {
cluster = new MiniDFSCluster.Builder(conf).numDataNodes(2)
.simulatedCapacities(capacities).build();
long initialCapacity = cluster.getNamesystem(0).getCapacityTotal();
assertTrue(initialCapacity > 0);
DataNode dn1 = cluster.getDataNodes().get(0);
DataNode dn2 = cluster.getDataNodes().get(1);
final DatanodeDescriptor dn2Desc = cluster.getNamesystem(0)
.getBlockManager().getDatanodeManager()
.getDatanode(dn2.getDatanodeId());
dn1.setHeartbeatsDisabledForTests(true);
cluster.setDataNodeDead(dn1.getDatanodeId());
assertEquals("Capacity shouldn't include DeadNode", dn2Desc.getCapacity(),
cluster.getNamesystem(0).getCapacityTotal());
assertEquals("NonDFS-used shouldn't include DeadNode",
dn2Desc.getNonDfsUsed(),
cluster.getNamesystem(0).getNonDfsUsedSpace());
// Wait for re-registration and heartbeat
dn1.setHeartbeatsDisabledForTests(false);
final DatanodeDescriptor dn1Desc = cluster.getNamesystem(0)
.getBlockManager().getDatanodeManager()
.getDatanode(dn1.getDatanodeId());
GenericTestUtils.waitFor(new Supplier<Boolean>() {
@Override public Boolean get() {
return dn1Desc.isAlive() && dn1Desc.isHeartbeatedSinceRegistration();
}
}, 100, 5000);
assertEquals("Capacity should be 0 after all DNs dead", initialCapacity,
cluster.getNamesystem(0).getCapacityTotal());
long nonDfsAfterReg = cluster.getNamesystem(0).getNonDfsUsedSpace();
assertEquals("NonDFS should include actual DN NonDFSUsed",
dn1Desc.getNonDfsUsed() + dn2Desc.getNonDfsUsed(), nonDfsAfterReg);
} finally {
if (cluster != null) {
cluster.shutdown();
}
}
}
}