DataStax Trouble with Creating Job Tracker

0
votes

I downloaded DataStax Enterprise and I was trying to run the following thier Portfolio Example (www.datastax.com/docs/datastax_enterprise2.2/solutions/dse_demo#dse-demo) but I got stuck. The issue I am having is that I can't get start the job tracker on a node.

I am using version 3.0.1 of dse on Ubuntu. I also have it set up with a single node, since I am just using one machine. I have already started a node and have both the opscenterd and opscenter-agent running.

When I run:

dse cassandra -t -j

I get the following output:

xss = -ea -javaagent:/usr/share/dse/cassandra/lib/jamm-0.2.5.jar -XX:+UseThreadPriorities -XX:ThreadPriorityPolicy=42 -Xms1024M -Xmx1024M -Xmn200M -XX:+HeapDumpOnOutOfMemoryError -Xss180k

Error: Exception thrown by the agent : java.rmi.server.ExportException: Port already in use: 7199; nested exception is: java.net.BindException: Address already in use

When try and use dsetool instead:

dsetool movejt localhost

I see, in log/cassandra/system.log, that if fails because:

Caused by: InvalidRequestException(why:Keyspace dse_system does not exist)

 at org.apache.cassandra.thrift.ThriftValidation.validateTable(ThriftValidation.java:83)
 at org.apache.cassandra.thrift.ThriftValidation.validateColumnFamily(ThriftValidation.java:135)
 at org.apache.cassandra.cql3.statements.SelectStatement$RawStatement.prepare(SelectStatement.java:973)
 at org.apache.cassandra.cql3.QueryProcessor.getStatement(QueryProcessor.java:243)
 at org.apache.cassandra.cql3.QueryProcessor.processInternal(QueryProcessor.java:129)
 at com.datastax.bdp.hadoop.mapred.TrackerManager.queryForJobTrackers(TrackerManager.java:202)
 at com.datastax.bdp.hadoop.mapred.TrackerManager.getJTNodes(TrackerManager.java:170)

What could be causing this? And how can it be fixed?

2
cassandradatastax-enterprise
can you restart your server? If, clear /var/lib/cassandra/data/ directory and then start the server - abhi

2 Answers

2
votes

My guess is that you have started up DSE as a plain C* node. To verify run: nodetool ring and dsetool ring and you will see the workload Cassandra for your node

The reason I believe this is because: 1) you have port in use error indicating you have a c* service already started 2) dse_system only gets created when you start up as a hadoop node, -t

To work around this, kill all java processes and try to restart with the -t option. It should detect that you dse_system has not been created and then create it. Then verify with dsetool ring.

0
votes

The above worked for me. Errors thrown were: 

$ dse pig
Error when fetching jobtracker address: 
Exception in thread "main" java.lang.NoClassDefFoundError: Could
$ dsetool jobtracker
Error when fetching jobtracker address: 
java.io.IOException: Could not find Job Tracker.

To resolve I did the following:

$ dse cassandra-stop
$ dse cassandra -t