[HDFS-8096] DatanodeMetrics#blocksReplicated will get incremented early and even for failed transfers

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 2.8.0, 3.0.0-alpha1
• Component/s: datanode
• Hadoop Flags: Reviewed
• Assignee: Vinayakumar B
• Reporter: Vinayakumar B
• Created: 08/Apr/15 10:26
• Updated: 30/Aug/16 01:37
• Resolved: 09/Apr/15 06:30

Description

    case DatanodeProtocol.DNA_TRANSFER:
      // Send a copy of a block to another datanode
      dn.transferBlocks(bcmd.getBlockPoolId(), bcmd.getBlocks(),
          bcmd.getTargets(), bcmd.getTargetStorageTypes());
      dn.metrics.incrBlocksReplicated(bcmd.getBlocks().length);

In the above code to handle replication transfers from namenode, DatanodeMetrics#blocksReplicated is getting incremented early, since the transfer will happen in background.
And even failed transfers also getting counted.

Correct place to increment this counter is DataTransfer#run().