kafka-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ben Ciceron <...@triggit.com>
Subject contrib/hadoop-consumer/README
Date Wed, 24 Aug 2011 21:58:16 GMT

in the README for the hadoop consumer , under the "HOW TO RUN"
section, steps 4 , what is the purpose of copy_jars ?
when i use /tmp/kafka/lib for ${hdfs.default.classpath.dir} on the
command line like you do in  test / test.properties then it copies the
jars under the local /tmp filesystem , not hdfs:///tmp , but if i
specified hdfs:///tmp/kafka/lib it complains it expects file:///.

same with th einput/output params in test / test.properties, which are
supposed to be hdfs path but the producer in step 3 of the same
section will write data in /tmp/kafka/data under the local /tmp
filesystem not hdfs://// .


View raw message