On Sun, Dec 9, 2012 at 9:33 AM, Jan Wieck <JanWieck at yahoo.com> wrote:
> On 12/6/2012 6:06 PM, Steve Singer wrote:
>> On 12-12-04 11:50 PM, David Rees wrote:
>>> On Tue, Dec 4, 2012 at 7:34 PM, Steve Singer <steve at ssinger.info> wrote:
>>> For example, table 1 is in set 1 and table 2 is in set 2 and table 1
>>> has a foreign key that refers to table 2.
>>>
>>> Changes that occur on the origin node 1 which will be the same for set
>>> 1/2 will obviously be fine, but is it possible for an event to be
>>> committed out-of-order on node 2?
>>
>> If you subscribe  set 2 first, then set 1 and if both set 1 and your not
>> using a different forwarder node, then I think you should be okay.
>
> The subscriptions should work either way since it is done under
> session_replication_role=replica, which suppresses foreign key checks.

Thanks!  I'll give it a shot and see how it goes.

-Dave