HDFS-16961. The blockReport RPC should throw UnregisteredNodeException when the storedDN is null

[ZanderXu]

Description of PR

The blockReport RPC should throw UnregisteredNodeException when the storedDN is null, the related code as bellows:

public void removeBRLeaseIfNeeded(final DatanodeID nodeID,
      final BlockReportContext context) throws IOException {
    namesystem.writeLock();
    DatanodeDescriptor node;
    try {
      // Here, if the node is null, we should throw UnregisteredNodeException insteat of NPE
      node = datanodeManager.getDatanode(nodeID);
      if (context != null) {
        if (context.getTotalRpcs() == context.getCurRpc() + 1) {
          long leaseId = this.getBlockReportLeaseManager().removeLease(node);
          BlockManagerFaultInjector.getInstance().
              removeBlockReportLease(node, leaseId);
          node.setLastBlockReportTime(now());
          node.setLastBlockReportMonotonic(Time.monotonicNow());
        }
        if (LOG.isDebugEnabled()) {
          LOG.debug("Processing RPC with index {} out of total {} RPCs in processReport 0x{}",
              context.getCurRpc(), context.getTotalRpcs(), Long.toHexString(context.getReportId()));
        }
      }
    } finally {
      namesystem.writeUnlock("removeBRLeaseIfNeeded");
    }
  }

[zhangshuyan0]

I found that when BlockManager#removeBRLeaseIfNeeded is called, BlockManager#checkBlockReportLease must have been called in advance. BlockManager#checkBlockReportLease already checked whether the nodeID exists in datanodeManager, is it necessary for us to check it again in BlockManager#removeBRLeaseIfNeeded?

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

Lines 2851 to 2859 in 67e02a9

	public boolean checkBlockReportLease(BlockReportContext context,
	final DatanodeID nodeID) throws UnregisteredNodeException {
	if (context == null) {
	return true;
	}
	DatanodeDescriptor node = datanodeManager.getDatanode(nodeID);
	if (node == null) {
	throw new UnregisteredNodeException(nodeID, null);
	}

[ZanderXu]

I found that when BlockManager#removeBRLeaseIfNeeded is called, BlockManager#checkBlockReportLease must have been called in advance. BlockManager#checkBlockReportLease already checked whether the nodeID exists in datanodeManager, is it necessary for us to check it again in BlockManager#removeBRLeaseIfNeeded?

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

Lines 2851 to 2859 in 67e02a9

	public boolean checkBlockReportLease(BlockReportContext context,
	final DatanodeID nodeID) throws UnregisteredNodeException {
	if (context == null) {
	return true;
	}
	DatanodeDescriptor node = datanodeManager.getDatanode(nodeID);
	if (node == null) {
	throw new UnregisteredNodeException(nodeID, null);
	}

@zhangshuyan0 Thanks for your review. I think we should check it again in BlockManager#removeBRLeaseIfNeeded, because there is a segment lock that bm.processReport() will unlock the write and BlockManager#removeBRLeaseIfNeeded hold the lock again. And there is a case that the datanode is removed after bm.processReport() and before BlockManager#removeBRLeaseIfNeeded.

[zhangshuyan0]

@ZanderXu Thanks for reply. It sounds like it makes sense. DatanodeDescriptor only can be removed from datanodeMap when registerDatanode() is called. I'm not sure if this is even possible to happen between checkBlockReportLease() and removeBRLeaseIfNeeded (). May need a UT to reproduce related problems.

[hadoop-yetus]

💔 -1 overall

Vote	Subsystem	Runtime	Logfile	Comment
+0 🆗	reexec	0m 56s		Docker mode activated.
			_ Prechecks _
+1 💚	dupname	0m 0s		No case conflicting files found.
+0 🆗	codespell	0m 1s		codespell was not available.
+0 🆗	detsecrets	0m 1s		detect-secrets was not available.
+1 💚	@author	0m 0s		The patch does not contain any @author tags.
-1 ❌	test4tests	0m 0s		The patch doesn't appear to include any new or modified tests. Please justify why no new tests are needed for this patch. Also please list what manual steps were performed to verify this patch.
			_ trunk Compile Tests _
+1 💚	mvninstall	41m 54s		trunk passed
+1 💚	compile	1m 28s		trunk passed with JDK Ubuntu-11.0.18+10-post-Ubuntu-0ubuntu120.04.1
+1 💚	compile	1m 19s		trunk passed with JDK Private Build-1.8.0_362-8u362-ga-0ubuntu1~20.04.1-b09
+1 💚	checkstyle	1m 6s		trunk passed
+1 💚	mvnsite	1m 30s		trunk passed
+1 💚	javadoc	1m 7s		trunk passed with JDK Ubuntu-11.0.18+10-post-Ubuntu-0ubuntu120.04.1
+1 💚	javadoc	1m 33s		trunk passed with JDK Private Build-1.8.0_362-8u362-ga-0ubuntu1~20.04.1-b09
+1 💚	spotbugs	3m 33s		trunk passed
+1 💚	shadedclient	25m 51s		branch has no errors when building and testing our client artifacts.
			_ Patch Compile Tests _
+1 💚	mvninstall	1m 19s		the patch passed
+1 💚	compile	1m 24s		the patch passed with JDK Ubuntu-11.0.18+10-post-Ubuntu-0ubuntu120.04.1
+1 💚	javac	1m 24s		the patch passed
+1 💚	compile	1m 14s		the patch passed with JDK Private Build-1.8.0_362-8u362-ga-0ubuntu1~20.04.1-b09
+1 💚	javac	1m 14s		the patch passed
+1 💚	blanks	0m 0s		The patch has no blanks issues.
+1 💚	checkstyle	0m 55s		the patch passed
+1 💚	mvnsite	1m 20s		the patch passed
+1 💚	javadoc	0m 53s		the patch passed with JDK Ubuntu-11.0.18+10-post-Ubuntu-0ubuntu120.04.1
+1 💚	javadoc	1m 27s		the patch passed with JDK Private Build-1.8.0_362-8u362-ga-0ubuntu1~20.04.1-b09
+1 💚	spotbugs	3m 27s		the patch passed
+1 💚	shadedclient	25m 30s		patch has no errors when building and testing our client artifacts.
			_ Other Tests _
-1 ❌	unit	248m 55s	/patch-unit-hadoop-hdfs-project_hadoop-hdfs.txt	hadoop-hdfs in the patch passed.
+1 💚	asflicense	0m 43s		The patch does not generate ASF License warnings.
		365m 7s

Reason	Tests
Failed junit tests	hadoop.hdfs.server.sps.TestExternalStoragePolicySatisfier
	hadoop.hdfs.server.datanode.TestDirectoryScanner

Subsystem	Report/Notes
Docker	ClientAPI=1.42 ServerAPI=1.42 base: https://ci-hadoop.apache.org/job/hadoop-multibranch/job/PR-5500/1/artifact/out/Dockerfile
GITHUB PR	#5500
Optional Tests	dupname asflicense compile javac javadoc mvninstall mvnsite unit shadedclient spotbugs checkstyle codespell detsecrets
uname	Linux 3cc2e9060108 4.15.0-206-generic #217-Ubuntu SMP Fri Feb 3 19:10:13 UTC 2023 x86_64 x86_64 x86_64 GNU/Linux
Build tool	maven
Personality	dev-support/bin/hadoop.sh
git revision	trunk / bb90232
Default Java	Private Build-1.8.0_362-8u362-ga-0ubuntu1~20.04.1-b09
Multi-JDK versions	/usr/lib/jvm/java-11-openjdk-amd64:Ubuntu-11.0.18+10-post-Ubuntu-0ubuntu120.04.1 /usr/lib/jvm/java-8-openjdk-amd64:Private Build-1.8.0_362-8u362-ga-0ubuntu1~20.04.1-b09
Test Results	https://ci-hadoop.apache.org/job/hadoop-multibranch/job/PR-5500/1/testReport/
Max. process+thread count	2430 (vs. ulimit of 5500)
modules	C: hadoop-hdfs-project/hadoop-hdfs U: hadoop-hdfs-project/hadoop-hdfs
Console output	https://ci-hadoop.apache.org/job/hadoop-multibranch/job/PR-5500/1/console
versions	git=2.25.1 maven=3.6.3 spotbugs=4.2.2
Powered by	Apache Yetus 0.14.0 https://yetus.apache.org

This message was automatically generated.

[Hexiaoqiao]

@ZanderXu Thanks for your works.

I'm not sure if this is even possible to happen between checkBlockReportLease() and removeBRLeaseIfNeeded().

+1, I am not sure if it is possible to happen after dig this code segment. Did you ever meet some NPE here?

[github-actions]

We're closing this stale PR because it has been open for 100 days with no activity. This isn't a judgement on the merit of the PR in any way. It's just a way of keeping the PR queue manageable.
If you feel like this was a mistake, or you would like to continue working on it, please feel free to re-open it and ask for a committer to remove the stale tag and review again.
Thanks all for your contribution.