Le mercredi 20 octobre 2021 11:40:18 CEST, vous avez écrit :
> > +# Setup the slot, and connect to it a first time
> > +$primary->run_log(
> > + [ 'pg_receivewal', '--slot', $slot_name, '--create-slot' ],
> > + 'creating a replication slot');
> > +$primary->psql('postgres',
> > + 'INSERT INTO test_table VALUES (generate_series(1,100));');
> > +$primary->psql('postgres', 'SELECT pg_switch_wal();');
> > +$nextlsn =
> > + $primary->safe_psql('postgres', 'SELECT pg_current_wal_insert_lsn();');
> > +chomp($nextlsn);
> > Wouldn't it be simpler to use CREATE_REPLICATION_SLOT with RESERVE_WAL
> > here, rather than going through pg_receivewal? It seems to me that
> > this would be cheaper without really impacting the coverage.
>
> You're right, we can skip two invocations of pg_receivewal like this (for
> the slot creation + for starting the slot a first time).
After sending the previous patch suite, I figured it would be worthwhile to
also have tests covering timeline switches, which was not covered before.
So please find attached a new version with an additional patch for those tests,
covering both "resume from last know archive" and "resume from the
replication slots position" cases.
--
Ronan Dunklau
