执行sqoop命令将数据从sqlserver中导入到hdfs,报错如下
20/04/28 14:30:22 ERROR tool.ImportTool: Imported Failed: We found column without column name.
Please verify that you've entered all column names in your query if using free form query
import (consider adding clause AS if you're using column transformation)
使用sqoop命令如下
sqoop import --connect ‘jdbc:sqlserver://localhost:1433;username=sa;password=1;database=personal_db’ --query “select pr_codeid_int,pr_name_vc,typecheck_vc,cast(sjc as bigint),cardcode from join_personal where $CONDITIONS” --target-dir /home/hadoop/hive_date/personal_db/join_personal -m 1 --split-by ‘1’ --fields-terminated-by ‘\001’
导致异常的原因是–query指定的sql语句中cast(sjc as bigint)未设置别名,指定别名即可cast(sjc as bigint) as sjc_b
sqoop语句参数含义
-m 1 //设置一个mapper,此参数要与 --split-by 联合使用
–split-by ‘1’ //设置mapper拆分依据的字段,正常情况下要使用select语句中的一个字段,
//但是我设置1个mapper,所以此处可以给一个常量即可
–fields-terminated-by ‘\001’ //设置数据导入到hdfs后的分隔符