Hi Farah,
The issue is that Hive doesn't know where to find Derby's data. In
build/dist/conf/hive-site.xml there is a property like this:
<property>
<name>javax.jdo.option.ConnectionURL</name>
<value>jdbc:derby:;databaseName=/var/lib/hive/metastore/${user.name}_db;create=true</value>
<description>JDBC connect string for a JDBC metastore</description>
</property>
You will want to change the value to point to your old Derby
directory. If you never configured a location for the Derby
directory, I believe it was created in the current working directory
where you invoked hive.
On Tue, Jun 22, 2010 at 1:15 PM, Omer, Farah <[email protected]> wrote:
> Hi,
>
>
>
> I have run into the following problem when I am trying to upgrade my current
> existing Hive to be Hive trunk. Till now, I was using the hive version that
> came in Cloudera’s VM, but now I need to run some tests using new SQL
> features that recently became available in Hive trunk.
>
> What I have done so far is that, I downloaded Hive trunk from apache SVN,
> and build it successfully, but the problem is when I run “./hive” from this
> new installation I don’t see any of those tables that I had previously
> created in my earlier Hive.
>
>
>
> Any pointers or suggestions on how to get all those old tables back, so that
> I can see them using the new Hive? I did copy over the entire data directory
> from /home/training/hive/data onto /home/training/hivetrunk/hive/data. Also
> the metastore is the default embedded database Derby.
>
>
>
> Thanks for any help with this.
>
>
>
> Farah Omer
>
> Senior DB Engineer, MicroStrategy, Inc.
>
>
>
> T: 703 2702230
>
> E: [email protected]
>
> http://www.microstrategy.com
>
>
>
>
>
>
--
Sarah Sproehnle
Educational Services
Cloudera, Inc
http://www.cloudera.com/training
