From: Hendrik Woltersdorf
Date: January 30 2013 12:02pm
Subject: Re: uneven data distribution, how to fix this?
List-Archive: http://lists.mysql.com/cluster/8481
Message-Id: <51090BD9.3060808@xcom.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Am 30.01.2013 11:37, schrieb umapathi b:
> restart the node 11 with ndbd --initial ..
> This should solve the problem ..I hope ..
>
>
>
> On Wed, Jan 30, 2013 at 3:59 PM, Hendrik Woltersdorf 
> <hendrik.woltersdorf@stripped <mailto:hendrik.woltersdorf@stripped>> wrote:
>
>     Hi,
>
>     in a test system we have a very uneven data distribution:
>
>     ndb_mgm> all report memory;
>     Node 10: Data usage is 55%(66989 32K pages of total 120000)
>     Node 10: Index usage is 10%(6657 8K pages of total 64032)
>     Node 11: Data usage is 14%(17221 32K pages of total 120000)
>     Node 11: Index usage is 10%(6654 8K pages of total 64032)
>
>     This started when one datanode was offline for a week due to
>     failing hardware.
>
>     How can I find out, which tables are affected and how can I fix this?
>
>     I tried:
>     - alter table ... reorganize partition
>     - optimize table ...
>     with the largest tables. But this did not change anything.
>
>

That did help. Thanks.

Just for the records: The cluster version is 7.2.8.