Support Questions

Find answers, ask questions, and share your expertise

Who agreed with this solution

avatar

Hi Wenbin,

 

I hope I understood well your use case. So you say that the data files are transferred to the correct HDFS location (with proper partitioning format directories, like partitionname=partitionvalue) but you want to make aware the Hive that there is a new partition on the HDFS.

In this case you need the

 

MSCK REPAIR TABLE table_name

 

command, please see:

https://cwiki.apache.org/confluence/display/Hive/LanguageManual+DDL#LanguageManualDDL-RecoverPartiti...)

In this case you don't need to execute ALTER TABLE ADD PARTITION for each new partition, Hive will recognize it.

 

In the newer Impala versions the same functionality exists in Impala as command:

ALTER TABLE table_name RECOVER PARTITIONS

 

Regards

 Miklos Szurap

Customer Operations Engineer

View solution in original post

Who agreed with this solution