[
https://issues.apache.org/jira/browse/DERBY-7084?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17153737#comment-17153737
]
Bryan Pendleton commented on DERBY-7084:
----------------------------------------
As a workaround, you can use ALTER TABLE ... RESTART WITH ... to set the
generator to the desired value after you import the data.
[https://builds.apache.org/job/Derby-docs/lastSuccessfulBuild/artifact/trunk/out/ref/rrefsqlj81859.html]
> Identity column data import does not increment generator
> --------------------------------------------------------
>
> Key: DERBY-7084
> URL: https://issues.apache.org/jira/browse/DERBY-7084
> Project: Derby
> Issue Type: Bug
> Reporter: Makkus B.
> Priority: Major
>
> When importing table data containing identity column vales (defined as
> GENERATED BY DEFAULT AS IDENTITY) via SYSCS_IMPORT_DATA (or
> SYSCS_IMPORT_TABLE) the data will be imported (when identity column was
> defined as GENERATED BY DEFAULT AS IDENTITY) , but the (internal) table index
> generator will not be updated. If the column is further defined as UNIQUE (or
> PRIMARY KEY) new data might not be added to the table (if the internal index
> will match an imported one).
> The expected behaviour would be that the internal index would always stay
> ahead of identity values inserted to the table. This is: for a new table the
> index defaults to 1. If data is imported to a new table, the index should not
> stay at 1 but become 1+max(imported_index). So the table can be continued to
> be used after the import.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
