https://community.cloudera.com/t5/Support-Questions/Solution-for-quot-slow-readprocessor-quot-warnings/m-p/122049#M84805 <P>I'm seeing the same behaviour appear half-way into a larger Pig job. </P><P>The HDFS datanodes seem to be in some sort of thrashing mode and I get infinite amount of log messages like:</P><P style="margin-left: 20px;">WARN datanode.DataNode (BlockReceiver.java:receivePacket(525)) - Slow BlockReceiver write packet to mirror took 5104ms (threshold=300ms)</P><P>The map/reduce jobs seem to not progress any more and leaving the job overnight does not give progress.</P> Wed, 06 Apr 2016 19:57:56 GMT eirik_ora 2016-04-06T19:57:56Z