hive-issue-inserting-records-to-partitioned-table

最新推荐文章于 2025-01-16 11:57:15 发布

转载最新推荐文章于 2025-01-16 11:57:15 发布 · 120 阅读

0 ·

CC 4.0 BY-SA版权

原文链接：http://www.cnblogs.com/suanec/p/10442433.html

在从HDP2.5.6升级到HDP2.6.4后，Hive在INSERT OVERWRITE TABLE操作上变得更加严格，导致错误。解决方法是在运行INSERT前删除数据并丢弃分区，或者不删除数据/丢弃分区，让INSERT操作直接替换。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

hive-issue-inserting-records-to-partitioned-table

Hi Sam,

Recently we upgraded our cluster from HDP2.5.6 to HDP2.6.4 and I am getting the similar error. With the current version Hive is more stricter on INSERT OVERWRITE TABLE. What it means is you might be deleting the data prior to loading the table and not dropping the partition when you do INSERT OVERWRITE TABLE.

To get around it,

Try to delete the data and drop partition,prior to running the INSERT OVERWRITE TABLE.

OR don't delete the data/drop partition for the external table let the INSERT OVERWRITE TABLE replace it.

Regards

Khaja Hussain.

Similar Error:

Caused by: java.util.concurrent.ExecutionException: org.apache.hadoop.hive.ql.metadata.HiveException: Destination directory hdfs://data_dir/pk_business=bsc/pk_data_source=pos/pk_frequency=bnw/pk_data_state=c13251_ps2111_bre000_pfc00000_spr000_pfs00000/pk_reporttype=BNN/pk_ppweek=2487 has not be cleaned up.

转载于:https://www.cnblogs.com/suanec/p/10442433.html