As mentioned in HDFS-12044, DataNode#transferBlock daemon is not calculated to xceiver count.
TestClientProtocolForPipelineRecovery#testZeroByteBlockRecovery fails on trunk
Mismatch between BlockManager#maxReplicationStreams and ErasureCodingWorker.stripedReconstructionPool pool size causes slow and bursty recovery
The patch simply puts DataTransfer to Daemon as the same in #transferBlock().
Thanks for the fix, Lei (Eddy) Xu.
The patch LGTM. +1 (non-binding).
+1 LGTM pending Jenkins, thanks Eddy!
This message was automatically generated.
Thanks a lot for the reviews, Hanisha Koneru and Andrew Wang
The test failures are not relevant. Passed locally on my laptop.
Committed to trunk and branch-2
SUCCESS: Integrated in Jenkins build Hadoop-trunk-Commit #12247 (See https://builds.apache.org/job/Hadoop-trunk-Commit/12247/)
HDFS-12215. DataNode.transferBlock does not create its daemon in the (lei: rev 36bada3032e438099ada9d865c3945d42c3e7c2a)