On Mon, 29 Jul 2019 at 00:09, Petr Jelinek <p...@2ndquadrant.com> wrote:


> Given that we don't process any other records in this function besides
> XLOG_HEAP2_MULTI_INSERT and XLOG_HEAP2_NEW_CID, it seems like simplest
> fix is to just remove the first check for fast forward and keep the one
> in XLOG_HEAP2_MULTI_INSERT.
>

Fix proposed by Petr, with comments as explained by Andres.

-- 
Simon Riggs                http://www.2ndQuadrant.com/
<http://www.2ndquadrant.com/>
PostgreSQL Solutions for the Enterprise

Attachment: allow_XLOG_HEAP2_NEW_CID_while_building_snapshot.v1.patch
Description: Binary data

Reply via email to