David Fetter wrote:
> 2) Is there another way to solve the problem of adding a foreign
> key constraint that points at a busy table?
Add it as NOT VALID and do a later VALIDATE CONSTRAINT step, after the
patch to reduce lock levels for ALTER TABLE goes in, perhaps?
--
Álvaro Herrera http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services