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

HDFS-16999. Fix wrong use of processFirstBlockReport(). (#5622). Contributed by Shuyan Zhang. 

Signed-off-by: Ayush Saxena <ayushsaxena@apache.org>
Signed-off-by: He Xiaoqiao <hexiaoqiao@apache.org>
This commit is contained in:
zhangshuyan 2023-05-09 10:05:28 +08:00 committed by GitHub
parent bdeca45294
commit 03bf8f982a
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
4 changed files with 66 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockManager.java
View File

@ -2913,7 +2913,7 @@ public class BlockManager implements BlockStatsMXBean {
return !node.hasStaleStorages();
}
if (storageInfo.getBlockReportCount() == 0) {
if (!storageInfo.hasReceivedBlockReport()) {
// The first block report can be processed a lot more efficiently than
// ordinary block reports. This shortens restart times.
blockLog.info("BLOCK* processReport 0x{} with lease ID 0x{}: Processing first "

9
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeStorageInfo.java
View File

@ -128,6 +128,10 @@ public class DatanodeStorageInfo {
/** The number of block reports received */
private int blockReportCount = 0;
/** Whether the NameNode has received block reports for this storage since it
* was started.*/
private boolean hasReceivedBlockReport = false;
/**
* Set to false on any NN failover, and reset to true
* whenever a block report is received.
@ -160,6 +164,10 @@ public class DatanodeStorageInfo {
return blockReportCount;
}
boolean hasReceivedBlockReport() {
return hasReceivedBlockReport;
}
void setBlockReportCount(int blockReportCount) {
this.blockReportCount = blockReportCount;
}
@ -188,6 +196,7 @@ public class DatanodeStorageInfo {
blockContentsStale = false;
}
blockReportCount++;
hasReceivedBlockReport = true;
}
@VisibleForTesting

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

@ -350,7 +350,7 @@ public class FsVolumeImpl implements FsVolumeSpi {
}
@VisibleForTesting
File getCurrentDir() {
public File getCurrentDir() {
return currentDir;
}

55
hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/blockmanagement/TestBlockManager.java
View File

@ -17,6 +17,9 @@
*/
package org.apache.hadoop.hdfs.server.blockmanagement;
import org.apache.hadoop.fs.BlockLocation;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileUtil;
import org.apache.hadoop.thirdparty.com.google.common.base.Joiner;
import org.apache.hadoop.thirdparty.com.google.common.collect.ImmutableList;
import org.apache.hadoop.thirdparty.com.google.common.collect.LinkedListMultimap;
@ -2066,4 +2069,56 @@ public class TestBlockManager {
// validateReconstructionWork return false, need to perform resetTargets().
assertNull(work.getTargets());
}
/**
* Test whether the first block report after DataNode restart is completely
* processed.
*/
@Test
public void testBlockReportAfterDataNodeRestart() throws Exception {
Configuration conf = new HdfsConfiguration();
try (MiniDFSCluster cluster = new MiniDFSCluster.Builder(conf)
.numDataNodes(3).storagesPerDatanode(1).build()) {
cluster.waitActive();
BlockManager blockManager = cluster.getNamesystem().getBlockManager();
DistributedFileSystem fs = cluster.getFileSystem();
final Path filePath = new Path("/tmp.txt");
final long fileLen = 1L;
DFSTestUtil.createFile(fs, filePath, fileLen, (short) 3, 1L);
DFSTestUtil.waitForReplication(fs, filePath, (short) 3, 60000);
ArrayList<DataNode> datanodes = cluster.getDataNodes();
assertEquals(datanodes.size(), 3);
// Stop RedundancyMonitor.
blockManager.setInitializedReplQueues(false);
// Delete the replica on the first datanode.
DataNode dn = datanodes.get(0);
int dnIpcPort = dn.getIpcPort();
File dnDir = dn.getFSDataset().getVolumeList().get(0).getCurrentDir();
String[] children = FileUtil.list(dnDir);
for (String s : children) {
if (!s.equals("VERSION")) {
FileUtil.fullyDeleteContents(new File(dnDir, s));
}
}
// The number of replicas is still 3 because the datanode has not sent
// a new block report.
FileStatus stat = fs.getFileStatus(filePath);
BlockLocation[] locs = fs.getFileBlockLocations(stat, 0, stat.getLen());
assertEquals(3, locs[0].getHosts().length);
// Restart the first datanode.
cluster.restartDataNode(0, true);
// Wait for the block report to be processed.
cluster.waitDatanodeFullyStarted(cluster.getDataNode(dnIpcPort), 10000);
cluster.waitFirstBRCompleted(0, 10000);
// The replica num should be 2.
locs = fs.getFileBlockLocations(stat, 0, stat.getLen());
assertEquals(2, locs[0].getHosts().length);
}
}
}