Steve Singer <ssinger at ca.afilias.info> writes:
> Jaime Casanova wrote:
>> On Thu, Mar 25, 2010 at 12:43 AM, Jaime Casanova
>
> Recall this was an issue where some inherited tables don't seem to have 
> any data in them after the initial set copy.
>
>
>>>> I'm trying to get a sense of if this is a problem that is being triggered by
>>>> some schemas with inheritance tables or if it is that some of your data is
>>>> causing things to fail (which still shouldn't happen)
>>>>
>>> yeah! i'm trying to figure out what caused the rollback if any
>>>
>> 
>> thinking a bit more on this, this should be as many rollbacks as
>> children has tcom_invitacion (one per month since 200803), and the log
>> shows that they were copied in 3 groups with other tables (that copied
>> fine) between those groups.
>> 
>> so, the one million question is: why "only" those tables?
>> i have another partitioned table but that was fine
>
> I think you will find that the difference is that the inherited table 
> has a lower slony table id (the id in the 'set add table') then the 
> master table.  The 'testinherit' test case in the regression tests 
> directory actually reproduces this (if one takes the time to look at the 
> output, which I should have been doing).
>
> However, this only seems to be a problem with 8.4, it works fine when I 
> run it against 8.3
>
>
>
> I have created bug 118 
> (http://www.slony.info/bugzilla/show_bug.cgi?id=118) to track this

Wow, finally a way of exposing some significance in the choice of table
ID numbers!

Not sure I want it to be significant, but it's finally something visible
:-).
-- 
"cbbrowne","@","ca.afilias.info"
Christopher Browne
"Bother,"  said Pooh,  "Eeyore, ready  two photon  torpedoes  and lock
phasers on the Heffalump, Piglet, meet me in transporter room three"