[
https://issues.apache.org/jira/browse/HIVE-1187?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12837563#action_12837563
]
Carl Steinbach commented on HIVE-1187:
--------------------------------------
@Ed: I don't think there is much overlap between this enhancement and what is
described in HIVE-1161. I expect that HIVE-1161 will involve triggering the
transfer of metadata over Thrift or JDBC from one metastore to another, or via
an agent that accesses both metastores via the client APIs. Dumping DDL from
one metastore and replaying it on another seems like a hacky way to
synchronize/transmit metadata.
> Implement ddldump utility for Hive Metastore
> --------------------------------------------
>
> Key: HIVE-1187
> URL: https://issues.apache.org/jira/browse/HIVE-1187
> Project: Hadoop Hive
> Issue Type: New Feature
> Components: Metastore
> Affects Versions: 0.6.0
> Reporter: Carl Steinbach
> Assignee: Carl Steinbach
>
> Implement a ddldump utility for the Hive metastore that will generate the QL
> DDL necessary to recreate the state of the current metastore on another
> metastore instance.
> A major use case for this utility is migrating a metastore from one database
> to another database, e.g. from an embedded Derby instanced to a MySQL
> instance.
> The ddldump utility should support the following features:
> * Ability to generate DDL for specific tables or all tables.
> * Ability to specify a table name prefix for the generated DDL, which will be
> useful for resolving table name conflicts.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
