HDFS-10625. VolumeScanner to report why a block is found bad. Contributed by Rushabh S Shah and Yiqun Lin.

Author: Yongjun Zhang

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/BlockSender.java

@@ -156,6 +156,9 @@ class BlockSender implements java.io.Closeable {
   /** The reference to the volume where the block is located */
   private FsVolumeReference volumeRef;
 
+  /** The replica of the block that is being read. */
+  private final Replica replica;
+
   // Cache-management related fields
   private final long readaheadLength;
 
@@ -238,7 +241,6 @@ class BlockSender implements java.io.Closeable {
             "If verifying checksum, currently must also send it.");
       }
 
-      final Replica replica;
       final long replicaVisibleLength;
       try(AutoCloseableLock lock = datanode.data.acquireDatasetLock()) {
         replica = getReplica(block, datanode);
@@ -688,8 +690,12 @@ public void verifyChecksum(final byte[] buf, final int dataOffset,
       checksum.update(buf, dOff, dLen);
       if (!checksum.compare(buf, cOff)) {
         long failedPos = offset + datalen - dLeft;
-        throw new ChecksumException("Checksum failed at " + failedPos,
-            failedPos);
+        StringBuilder replicaInfoString = new StringBuilder();
+        if (replica != null) {
+          replicaInfoString.append(" for replica: " + replica.toString());
+        }
+        throw new ChecksumException("Checksum failed at " + failedPos
+            + replicaInfoString, failedPos);
       }
       dLeft -= dLen;
       dOff += dLen;

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/VolumeScanner.java

@@ -281,12 +281,13 @@ public void handle(ExtendedBlock block, IOException e) {
             volume.getBasePath(), block);
         return;
       }
-      LOG.warn("Reporting bad {} on {}", block, volume.getBasePath());
+      LOG.warn("Reporting bad " + block + " with volume "
+          + volume.getBasePath(), e);
       try {
         scanner.datanode.reportBadBlocks(block, volume);
       } catch (IOException ie) {
         // This is bad, but not bad enough to shut down the scanner.
-        LOG.warn("Cannot report bad " + block.getBlockId(), e);
+        LOG.warn("Cannot report bad block " + block, ie);
       }
     }
   }