Konstantin Knizhnik <k(dot)knizhnik(at)postgrespro(dot)ru> writes:
> I think that using restart_lsn instead of confirmed_flush is not right approach.
> If restart_lsn is not available and confirmed_flush is pointing to page
> boundary, then in any case we should somehow handle this case and adjust
> startlsn to point on the valid record position (by jjust adding page header
> size?).
Well, restart_lsn is always available on live slot: it is initially set
in ReplicationSlotReserveWal during slot creation.
--
Arseny Sher
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company