On Fri, Sep 11, 2020 at 06:42:09PM +0300, Anastasia Lubennikova wrote:
> I agree that transactional update in index_set_state_flags() is good for
> both cases, that you mentioned and don't see any restrictions. It seems that
> not doing this before was just a loose end, as the comment from 813fb03
On 03.09.2020 11:04, Michael Paquier wrote:
Hi all,
For a couple of things I looked at lately, it would be really useful
to make index_state_set_flags() transactional and replace its use of
heap_inplace_update() by CatalogTupleUpdate():
- When dropping an index used in a replica identity, we cou
