Hi Stack: Thanks for the reply. Yes. I set the full-path to hbase/conf.
Actually, I printed out all environmental variables in my map/reduce program, and I can see HADOOP_PATH is set to the right place (to hbase jar, and hbase conf directory) Just want to double check, if everything works fine, my map/reduce program should look for <master_machine>:60000 instead of localhost:60000, right? Regards tp On Wed, Sep 9, 2009 at 6:45 AM, stack <[email protected]> wrote: > On Tue, Sep 8, 2009 at 6:13 PM, charles du <[email protected]> wrote: > > > Hi: > > > > I installed hbase 0.19.3 and hadoop 0.19.1. I tried to run the BulkImport > > example on http://wiki.apache.org/hadoop/Hbase/MapReduce, get the > > following > > error. > > > > "org.apache.hadoop.hbase.MasterNotRunningException: > localhost:60000" > > > > From the error message, it looks like the hadoop looks at the wrong place > > for the hbase configuration. I added paths to hbase-0.19.3.jar, > > hbase-0.19.3-test.jar, hbase/conf to ""export HADOOP_CLASSPATH =" in > > 'hadoop-env.sh', populated the change to every hadoop machine, and > > restarted > > hadoop. > > > > > It looks like you are doing all the right stuff and I'd agree with your > speculation, that its not finding the hbase configuration. You have set > full paths to hbase/conf directory in HADOOP_CLASSPATH? > > St.Ack > -- tp
