It is unlikely to be specifically caused by where clause. Are you able to run this query using hive cli ? Are you able to run any query that involves running a MR job through jdbc ?
What do you see in the hive logs ? On Tue, Jul 16, 2013 at 1:10 AM, ch huang <justlo...@gmail.com> wrote: > here is my test output,why the beeline report error when i use where > condition?? > > hive> select foo from demo_hive where bar='value3'; > Total MapReduce jobs = 1 > Launching Job 1 out of 1 > Number of reduce tasks is set to 0 since there's no reduce operator > Starting Job = job_1373509276088_0015, Tracking URL = > http://CH22:8088/proxy/application_1373509276088_0015/ > Kill Command = /usr/lib/hadoop/bin/hadoop job -kill job_1373509276088_0015 > Hadoop job information for Stage-1: number of mappers: 1; number of > reducers: 0 > 2013-07-16 16:00:05,538 Stage-1 map = 0%, reduce = 0% > 2013-07-16 16:00:11,958 Stage-1 map = 100%, reduce = 0%, Cumulative CPU 5.6 > sec > MapReduce Total cumulative CPU time: 5 seconds 600 msec > Ended Job = job_1373509276088_0015 > MapReduce Jobs Launched: > Job 0: Map: 1 Cumulative CPU: 5.6 sec HDFS Read: 245 HDFS Write: 2 > SUCCESS > Total MapReduce CPU Time Spent: 5 seconds 600 msec > OK > 3 > Time taken: 15.236 seconds > > 0: jdbc:hive2://192.168.10.22:10000> select foo from demo_hive where > bar='value3'; > Error: Error while processing statement: FAILED: Execution Error, return > code 1 from org.apache.hadoop.hive.ql.exec.MapRedTask (state=08S01,code=1) > Error: Error while processing statement: FAILED: Execution Error, return > code 1 from org.apache.hadoop.hive.ql.exec.MapRedTask (state=08S01,code=1)
