About Copying Entire Databases - Teradata Data Mover

Teradata Data Mover User Guide

Product
Teradata Data Mover
Release Number
16.00
Published
December 2016
Language
English (United States)
Last Update
2018-03-29
dita:mapPath
rdo1467305237457.ditamap
dita:ditavalPath
ft:empty
dita:id
B035-4101
lifecycle
previous
Product Category
Analytical Ecosystem
Using Teradata ARC, you can copy all the objects from a source to a target database without specifying each object in the database. This is a useful alternative to specifying a copy of individual tables, especially when the database has a large number of objects. When copying the entire database to a target, you can do any or all of the following:
  • Exclude specific tables from being copied
  • Rename tables to be copied in the target
  • Copy partial table data
  • Copy join and hash indexes
  • Copy triggers
  • Copy permanent journals
  • Enable the compare DDL feature for all tables to be copied
  • Disable the compare DDL feature for selected tables to be copied
  • Rename and relocate hash and join indexes
  • Specify a target hash index or join index
  • Specify a target index database
  • Copy entire databases and all their children
  • Exclude child databases from being copied
  • Rename tables in a child database
  • Copy specific tables in a child database
  • Copy to a different target database
  • Copy a parent database with its children, but relocate a child database to a different database on the target

You can only copy an entire database to a target database in a different Teradata Database system. The Teradata ARC utility must be used. Teradata PT API operators and Teradata JDBC utility do not support copying an entire database. The specified target database must exist and have adequate space.

Databases cannot be copied between Teradata and Aster or Hadoop systems.

When using ARC to copy data to a target system that is one or more major releases newer than the source system, the job log may contain the following ARC warning:
Warning ARC1257: A cross-release data restore requires that a Teradata SQL conversion script be run.
Data Mover does not execute this post-data restore conversion script automatically. In most cases, it is not necessary to execute this script unless you are copying an entire database where the database contains an external stored procedure (XSP) or a user-defined function (UDF).
When using Teradata ARC to copy an entire database on which statistics have been collected, Teradata ARC copies the statistics of all objects within the database, even if the copyStats attribute was false or not specified. When the database being moved contains a join index that has statistics, there is a known issue that will cause the Data Mover job to fail. By default, Data Mover does not move join indices when copying entire databases with ARC. This is to avoid issues where the join index may reference tables in other databases that are not present on the target system. Data Mover uses the SKIP JOIN INDEX ARC option to tell ARC not to move any join indices in the database. When moving statistics for a database, ARC does not have any method for not moving the statistics that apply to the join indices in the database; therefore, ARC will move the entire database without the join indices and get an error when trying to move the statistics that apply to the join indices. The workarounds for this restriction are as follows:
  • Enable the database.copy.joinIndex configuration property. In this case, Data Mover will not use the SKIP JOIN INDEX clause when calling ARC. ARC will move the join indices in the database before moving the statistics. Enabling this configuration property will affect all Data Mover jobs, which may not be desirable if the join indices should not be moved with full database copies.
  • Drop the statistics on the join index prior to using Data Mover to copy the entire database; then, recollect the statistics on the join index afterwards. In this case, any previously collected statistics on the join index will be lost, which may not be a desirable option.
Choosing to copy the join index separately from the database will not resolve the issue. Data Mover will copy the join index only after the ARC job has completed copying the entire database and its statistics. The join index will still be missing when ARC attempts to copy the statistics.