AnsweredAssumed Answered

AsyncHbaseSink tries to communicate with zookeeper on TCP port 2181 instead of 5181

Question asked by kkuehl on Apr 4, 2013
This question seems directly related to

I am using the mapr m3 vmware image.
/opt/mapr/hbase/hbase-0.92.1/conf/hbase-site.xml contains
and netstat -an | grep LISTEN confirms zookeeper is running on 5181

I have written a prototype asynchbasesink for flume.

I have attempted to add /opt/mapr/hbase/hbase-0.92.1/conf to the beginning of the classpath as recommended in the answers to the similar question above.
flume-ng agent --classpath /opt/mapr/hbase/hbase-0.92.1/conf/ --conf $CWD/conf --conf-file $CWD/conf/flume.conf -Dflume.root.logger=DEBUG,console,LOGFILE -n agent

When the above command line is executed, the following is displayed in the console:
 + exec /usr/lib/jvm/java-6-sun- -Xms100m -Xmx200m -Dflume.root.logger=DEBUG,console,LOGFILE -cp '/opt/mapr/flume/flume-1.3.1/lib/*:/opt/mapr/hbase/hbase-0.92.1/conf:lib/*:ct-flume-agent/target/*:

It still tries to connect to tcp port 2181:
2013-04-04 08:18:11,066 (lifecycleSupervisor-1-0-SendThread(localhost:2181)) [INFO - org.apache.zookeeper.ClientCnxn$SendThread.logStartConnect(] Opening socket connection to server localhost/ Will not attempt to authenticate using SASL (Unable to locate a login configuration)
2013-04-04 08:18:11,070 (lifecycleSupervisor-1-0-SendThread(localhost:2181)) [WARN - org.apache.zookeeper.ClientCnxn$] Session 0x0 for server null, unexpected error, closing socket connection and attempting reconnect Connection refused

I also have tried adding the folloging to my derived AsyncHbaseEventSerializer class's initialize()
Configuration conf = HBaseConfiguration.create();
conf.setInt("", 5181);
with identical results.

Any help/insight would be appreciated.