I am still confused:

(1) What I observed is that not only the first msg got wrong partition
number, but also
     the second message.

(2) If kafka uses the first msg to get the correct msg. Does it mean that
if we use the
customized partitioner, we cannot get the right partition for this message?

On Fri, Jan 11, 2013 at 9:06 PM, Jun Rao <[EMAIL PROTECTED]> wrote:
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB