Oct
29
2015
SQL // Hadoop

Uniqueidentifier data type in SQL Server not supported in Hive

Recently, while using Sqoop to pull data into Hadoop from MS SQL Server, I found an issue with a table whose primary key was a uniqueidentifer column (GUID).  The problem was nicely documented on StackOverflow here.

As a result I’ve taken the approach of designing the data load routines in Java to alternate between using –split-by and –num-mappers depending the table schemas and it is working beautifully.

Calendar

<<  October 2017  >>
MoTuWeThFrSaSu
2526272829301
2345678
9101112131415
16171819202122
23242526272829
303112345

View posts in large calendar

Page List

    RecentComments

    None