hadoop - issue with hive partitioning and bucketing in CDH 5.10 quick VM -


i new area , got stuck in simple issue.

i loading data hive table (using insert command table tset1) partitioned udate , day bucket.

insert overwrite test1 partition(udate) select id,value,udate,day tset1; 

so issue when loading data taking wrong value in partition column. day taken partition because in table last column during data load it's taking day udate.

how can force query take right value during data load?

hive (testdb)> create table test1_buk(id int, value string,day int) partitioned by(udate string) clustered by(day) 5 buckets row format delimited fields terminated ',' stored textfile;  hive (testdb)> desc tset1; ok col_name    data_type   comment id                      int                                          value                   string                                       udate                   string                                       day                     int   hive (testdb)> desc test1_buk; ok col_name    data_type   comment id                      int                                          value                   string                                       day                     int                                          udate                   string                                        # partition information       # col_name              data_type               comment               udate                   string   hive (testdb)> select * test1_buk limit 1; ok test1_buk.id    test1_buk.value test1_buk.day   test1_buk.udate 5   w   2000    10 

please help.


Comments

Popular posts from this blog

angular - Ionic slides - dynamically add slides before and after -

minify - Minimizing css files -

Add a dynamic header in angular 2 http provider -