Are you using CombineHiveInputFormat ? And are you using compressed text files ?
Thanks, -namit From: 朱鹤群 [mailto:[email protected]] Sent: Monday, April 12, 2010 12:53 AM To: [email protected] Subject: Problem about 'like' condition Hi, Has anybody got problem when query in hive with condition (field like '%string%')? I got problem that the return records are duplicated. My hadoop cluster has 1 namenode, 1 jobtracker and 3 slaves. On each slave, there is 2 mappers and 2 reducers. All data are kept in gz file. In this case, the record duplicated for 15 times. Will anybody give some clue? Thanks. -- Best Regards, Gary Zhu
