openjpa-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Donald Woods (JIRA)" <j...@apache.org>
Subject [jira] Commented: (OPENJPA-1812) Schema synchronizer fails to build tables and indexes when running with H2 in-memory database
Date Thu, 23 Sep 2010 13:11:36 GMT

    [ https://issues.apache.org/jira/browse/OPENJPA-1812?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12914022#action_12914022
] 

Donald Woods commented on OPENJPA-1812:
---------------------------------------

Can you post the output from when the disk based H2 db is used?  Sounds like a H2 problem
with in-memory behaving differently than on-disk, but lets look at the logs before declaring
that....
BTW - we run Derby as both on-disk (-Ptest-derby) and in-memory (-Ptest-derbymem) and all
the junits pass.


> Schema synchronizer fails to build tables and indexes when running with H2 in-memory
database
> ---------------------------------------------------------------------------------------------
>
>                 Key: OPENJPA-1812
>                 URL: https://issues.apache.org/jira/browse/OPENJPA-1812
>             Project: OpenJPA
>          Issue Type: Bug
>    Affects Versions: 2.0.1
>            Reporter: Willis Blackburn
>            Priority: Minor
>
> Using the example from OPENJPA-1804 I added HostName and EmailAddress to the class list,
updated the JVM launcher command to use the Java agent enhancer, and added:
>             <property name="openjpa.jdbc.SynchronizeMappings" value="buildSchema(foreignKeys=true)"/>
> Note also:
>             <property name="openjpa.ConnectionURL" value="jdbc:h2:mem:test"/>
> When I tried to run the example, it failed when attempting to create an index, saying
that address_book_table (which it created a few milliseconds earlier) did not exist.
> Changing H2 to use a database on disk fixed the problem:
>             <property name="openjpa.ConnectionURL" value="jdbc:h2:~/openjpatestdb"/>
> But why is OpenJPA sensitive to whether I'm using an in-memory or disk H2 database?
> I tried opening the same in-memory database URL using my IDE's JDBC console.  Of course
I got a different database, since the IDE is running in its own JVM.  But I was able to execute
the "create table" from the OpenJPA log followed by the "create index" without any trouble.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message