McDonald,   I get the same error even in the Cloudera quick start VM 5.3   As you know, it is a pseudo-distributed hadoop cluster, meaning, all services including HiveServer2 are on one machine only.   So it seems more like a beeline bug, rather than the explanation you provided.   Same works with the hive CLI, but fails with beeline on Cloudera 5.3 Quick start VM   beeline   0: jdbc:hive2://localhost:10000> LOAD DATA LOCAL INPATH '/home/cloudera/datasets/ml-100k/u.data' INTO TABLE u_data; Error: Error while compiling statement: FAILED: SemanticException Line 1:23 Invalid path ''/home/cloudera/datasets/ml-100k/u.data'': No files matching path file:/home/cloudera/datasets/ml-100k/u.data (state=42000,code=40000) 0: jdbc:hive2://localhost:10000> !quit Closing: 0: jdbc:hive2://localhost:10000   The /home/cloudera/datasets/ml-100k/u.data was owned by cloudera:cloudera. I even gave chmod o+r  to this file. But still same problems.     hive   Logging initialized using configuration in file:/etc/hive/conf.dist/hive-log4j.properties hive> LOAD DATA LOCAL INPATH '/home/cloudera/datasets/ml-100k/u.data' INTO TABLE u_data; Copying data from file:/home/cloudera/datasets/ml-100k/u.data Copying file: file:/home/cloudera/datasets/ml-100k/u.data Loading data to table default.u_data Table default.u_data stats: [numFiles=1, numRows=0, totalSize=1979173, rawDataSize=0] OK Time taken: 3.136 seconds hive>   -Thanks Srini ... View more