[Postgres-xl-general] Adding new datanode and "ALTER TABLE REDISTRIBUTE"

Pavan Deolasee pavan.deolasee at gmail.com
Tue Sep 16 10:35:34 PDT 2014


On Tue, Sep 16, 2014 at 10:57 PM, Dennis <dennisr at visi.com> wrote:

> Hi all,   I am confused by the current documentation.  I have just added
> new datanode to my cluster and now I need to redistribute the tables to the
> new node.  The documentation says I need to "Redistribute existing data
> by using ALTER TABLE REDISTRIBUTE. The new datanode is now ready.”
>  (From: http://files.postgres-xl.org/documentation/add-node-datanode.html)
>
> But there doesn’t seem to be a “ATLER TABLE REDISTRIBUTE” command.  So I
> am wondering if the correct command is to issue an “ALTER TABLE foo ADD
> NODE(new_node);” command and if that will redistribute the table and it’s
> indexes.  I am using both replicated and round robin distribution schemes
> currently.
>
>
AFAIK “ALTER TABLE foo ADD NODE(new_node);” is the correct way to
redistribute data. You can use ALTER TABLE foo DISTRIBUTE BY <scheme> to
change the distribute scheme itself (e.g. changing distribution column or
type). But to redistribute data to a new set of nodes, one should use
ADD/DELETE NODE mechanism.

I would also suggest using the latest sources from the git repository since
we have fixed couple of bugs in this area.

Thanks,
Pavan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.postgres-xl.org/private.cgi/postgres-xl-general-postgres-xl.org/attachments/20140916/95deed53/attachment.htm>


More information about the postgres-xl-general mailing list