Managed to work it out. Once I had hbase-site.xml in the Hadoop conf/ directory, I just had to restart hbase AND hadoop to get things working. I had restarted hadoop but it seems hbase was not running. My mistake, sorry.
Thanks for your help, St.Ack! John -----Original Message----- From: John I. Noecker Jr. [mailto:[email protected]] Sent: Tuesday, July 07, 2009 9:51 AM To: [email protected] Subject: RE: HBase Error I have tried adding hbase-site.xml to the HADOOP_CLASSPATH in hadoop-env.sh on each of the nodes, and have also tried putting hbase-site.xml into the Hadoop conf/ directory, but I still get that they are trying (unsuccessfully) to connect to localhost, rather than elroy. Using HBase from the shell works fine. John -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of stack Sent: Monday, July 06, 2009 7:39 PM To: [email protected] Subject: Re: HBase Error You have added the hbase-site.xml where? Is it in the hadoop conf directory or have you added the hbase conf directory to HADOOP_CLASSPATH in hadoop-env.sh? Or did you build the hbase-site.xml into your job jar? St.Ack On Mon, Jul 6, 2009 at 3:41 PM, John I. Noecker Jr. <[email protected] > wrote: > Hello all, > > > > I am using HBase within a Map/Reduce job, and when I run the job, it > exits with the following error: > > > > 09/07/06 18:34:42 INFO mapred.JobClient: Task Id : > attempt_200907061758_0004_m_000000_0, Status : FAILED > > org.apache.hadoop.hbase.MasterNotRunningException: localhost:60000 > > at > org.apache.hadoop.hbase.client.HConnectionManager$TableServers.getMaster > (HConnectionManager.java:239) > > at > org.apache.hadoop.hbase.client.HConnectionManager$TableServers.locateReg > ion(HConnectionManager.java:439) > > at org.apache.hadoop.hbase.client.HTable.<init>(HTable.java:115) > > at org.apache.hadoop.hbase.client.HTable.<init>(HTable.java:98) > > at com.proteus.athena.hadoop.xml.XmlMapper.map(XmlMapper.java:74) > > at com.proteus.athena.hadoop.xml.XmlMapper.map(XmlMapper.java:1) > > at org.apache.hadoop.mapred.MapRunner.run(MapRunner.java:50) > > at org.apache.hadoop.mapred.MapTask.run(MapTask.java:342) > > at org.apache.hadoop.mapred.Child.main(Child.java:158) > > > > I am running Hadoop 0.19.1 with HBase 0.19.3. I have a cluster of 5 > nodes, with one node (named 'elroy' == 10.1.0.200) serving as the Hadoop > namenode and jobtracker and the HBase Master. > > > > My hbase-site.xml: > > <configuration> > > <property> > > <name>hbase.rootdir</name> > > <value>hdfs://elroy/hbase</value> > > </property> > > > > <property> > > <name>hbase.master</name> > > <value>elroy:6000</value> > > </property> > > </configuration> > > > > My regionservers: > > 10.1.0.200 > > 10.1.0.201 > > 10.1.0.202 > > 10.1.0.203 > > 10.1.0.204 > > > > It seems that when the Map/Reduce job attempts to contact the HBase > database, it tries to do so on localhost:6000 instead of elroy:6000... > Since I have hbase.master set in hbase-site.xml, I am not sure why it is > doing this. > > > > Any suggestions? > > > > Thanks, > > John Noecker Jr. > >
