On 2016-04-18 20:43:30 +0900, Michael Paquier wrote: > Yeah, introducing a new WAL record to address this issue in > back-branches would not be an issue, and that's what we should do. For > HEAD, let's add that in the commit record.
I'm not sure why/how you'd do it that way in HEAD? I mean the only reason not to use a separate record is the standby incompatibility. - Andres -- Sent via pgsql-hackers mailing list (email@example.com) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers