Re: Logical Replication - behavior of ALTER PUBLICATION .. DROP TABLE and ALTER SUBSCRIPTION .. REFRESH PUBLICATION

On Sat, Jan 16, 2021 at 6:08 PM Bharath Rupireddy
<bharath(dot)rupireddyforpostgres(at)gmail(dot)com> wrote:
>
> On Sat, Jan 16, 2021 at 12:21 PM Bharath Rupireddy
> <bharath(dot)rupireddyforpostgres(at)gmail(dot)com> wrote:
> > > In the test, can we have multiple publications for the subscription
> > > because that is how we discovered that the originally proposed patch
> > > was not correct? Also, is it possible to extend some of the existing
> > > tests in 001_rep_changes.pl or anywhere so that we can avoid some cost
> > > of the replication setup.
> >
> > Sure, I will add the multiple publications use case provided by japin
> > and also see if I can move the tests from 100_bugs.pl to
> > 0001_rep_changes.pl.
>
> Attaching v5 patch set. 0001 - No change. 0002 - I moved the tests to
> 001_rep_changes.pl so that we could avoid creation of subscriber,
> publisher nodes and other set up.
>

Thanks for the patch. Few comments:

+
+# Test replication with multiple publications for subscription
+

While checking the existing tests in 001_rep_changes.pl, I came across
the below test which has multiple publications, won't that suffice the
need?

"CREATE SUBSCRIPTION tap_sub CONNECTION '$publisher_connstr'
PUBLICATION tap_pub, tap_pub_ins_only"

BTW, have we tried to check if this problem exists in back-branches?
It seems to me the problem has been recently introduced by commit
69bd60672a. I am telling this by looking at code and have yet not
performed any testing so I could be wrong.

--
With Regards,
Amit Kapila.