Horton

Re: More tests with USING INDEX replident and dropped indexes

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: Rahila Syed <rahila.syed@2ndquadrant.com>
Cc: Euler Taveira <euler.taveira@2ndquadrant.com>, Masahiko Sawada <masahiko.sawada@2ndquadrant.com>, Postgres hackers <pgsql-hackers@lists.postgresql.org>
Date: 2020-09-02T06:00:44Z
Lists: pgsql-hackers 
On Wed, Sep 02, 2020 at 09:27:33AM +0530, Rahila Syed wrote:
> TBH, I am not sure.  I think it is a reasonable change. It is even
> indicated in the
> comment above index_set_state_flags() that it can be made transactional.
> At the same time, probably we can just go ahead with current
> inconsistent update of relisreplident and indisvalid flags. Can't see what
> will break with that.

Yeah, that's true as well.  Still, I would like to see first if people
are fine with changing this code path to be transactional.  This way,
we will have zero history in the tree where there was a risk for an
inconsistent window.
--
Michael

Commits

  1. Add regression tests for REPLICA IDENTITY with dropped indexes