Unable to rename Hive Table Partition

to create new topics or reply. | New User Registration

This topic contains 4 replies, has 2 voices, and was last updated by  Dhirendra Singh 1 year, 3 months ago.

  • Creator
  • #51219

    Dhirendra Singh

    I am trying to rename a Partition using following command:
    ALTER TABLE dummy_table PARTITION (year=’2014′, month=’01’, day=’01’) RENAME TO PARTITION (year=’2014′, month=’01’, day=’02’);
    This runs fine on my single node unsecured cluster. However, when I run same command with same set of data-sets on a Kerberos secured multi-node cluster, it gives following error:
    FAILED: Execution Error, return code 1 from org.apache.hadoop.hive.ql.exec.DDLTask. Unable to rename partition.
    Even the hive logs also have only this much of log. I am not able to understand the root cause of this issue.

    Here is the table partition information
    # col_name data_type comment
    year int None
    month int None
    day int None

    Please help me understand the root-cause of this problem and resolve it.

Viewing 4 replies - 1 through 4 (of 4 total)

You must be to reply to this topic. | Create Account

  • Author
  • #51502

    Dhirendra Singh

    Got the problem. It seems that renaming a partition is not allowed for an internal table. I am able to rename the partition for an external table but it does not work for an internal table.


    Dhirendra Singh

    Moreover can anyone tell me where can I find the source code for hive libraries so that I can debug this issue? I am using hortonworks hive version I have checked the hortonworks maven repository but it is not there.


    Dhirendra Singh

    No Thejas, actually all other commands are working fine like create partition, drop partition, show partitions etc. Even I am able to change the partition fileformat and location as well. This error comes only when I am trying to rename the partition.

    Thanks for your reply!!!


    Thejas Nair

    Are you using a remote metastore server ? Is there any error in log of metastore server?

Viewing 4 replies - 1 through 4 (of 4 total)
Hortonworks Data Platform
The Hortonworks Data Platform is a 100% open source distribution of Apache Hadoop that is truly enterprise grade having been built, tested and hardened with enterprise rigor.
Get started with Sandbox
Hortonworks Sandbox is a self-contained virtual machine with Apache Hadoop pre-configured alongside a set of hands-on, step-by-step Hadoop tutorials.
Modern Data Architecture
Tackle the challenges of big data. Hadoop integrates with existing EDW, RDBMS and MPP systems to deliver lower cost, higher capacity infrastructure.