Details
-
Bug
-
Status: Open
-
Major
-
Resolution: Unresolved
-
0.11.0
-
None
-
None
-
ubuntu12.04,hive-0.11.0,hadoop1.02
Description
set hive.groupby.skewindata=true;
select count(distinct uid) as uid,type from test
where dt='20130818
group by type;
just one job.not merge the middle result.
the output like:
100 1
200 2
300 1
400 2
it should be:
400 1
600 2