The reason for this error is that a delimiter error is used between table fields in the specified hive for Sqoop read parsing to be incorrect. If the result of the MapReduce operation Rollup is performed by hive, the default delimiter is ' \001 ', otherwise the delimiter should be ' \ t ' if imported from an HDFs file. Here I am the result of hive performing the MapReduce analysis rollup, so the default separation is ' \001 '. The Sqoop command modifies the following, specifying the delimiter:
[Email protected] ~]$ sqoop-export--connect jdbc:mysql://secondmgt:3306/transport--username hive--password Hive --table jtxx--export-dir/hive/warehouse/hive.db/temptrans--input-fields-terminated-by ' \001 ';
For other questions see: http://blog.csdn.net/niityzu/article/details/45190787
Problems with importing tables from hive to MySQL using Sqoop