> I would like to know the ctid's of the -1 rows in the copied table,
> along with the ctid's of the rows they share magic_ids with, and
> the ctid's of the rows with those same magic_ids in the original.
> I'm wondering whether the affected rows are physically clustered ...
i tried:
create table qqq as select cmax as o_cmax, xmax as o_xmax, cmin as
o_cmin, xmin as o_xmin, ctid as o_ctid, * from sssssss.xobjects;
but the resulting table didn't have -1 values:
$ select xobject_id, count(*) from qqq group by 1 having count(*) > 1;
xobject_id | count
------------+-------
(0 rows)
$ select o_cmax,o_xmax,o_cmin,o_xmin,o_ctid, xobject_id, order_id from qqq where xobject_id = -1;
o_cmax | o_xmax | o_cmin | o_xmin | o_ctid | xobject_id | order_id
--------+--------+--------+--------+--------+------------+----------
(0 rows)
i'm checking now something else, but later will try to get better grasp on those bad rows.
Best regards,
depesz
--
The best thing about modern society is how easy it is to avoid contact with it.
http://depesz.com/