Re: Minimal logical decoding on standbys

On Wed, Apr 5, 2023 at 6:14 PM Drouvot, Bertrand
<bertranddrouvot(dot)pg(at)gmail(dot)com> wrote:
>
> On 4/5/23 12:28 PM, Amit Kapila wrote:
> > On Wed, Apr 5, 2023 at 2:41 PM Drouvot, Bertrand
> > <bertranddrouvot(dot)pg(at)gmail(dot)com> wrote:
> >> Maybe we could change the doc with something among those lines instead?
> >>
> >> "
> >> Existing logical slots on standby also get invalidated if wal_level on primary is reduced to
> >> less than 'logical'. This is done as soon as the standby detects such a change in the WAL stream.
> >>
> >> It means, that for walsenders that are lagging (if any), some WAL records up to the parameter change on the
> >> primary won't be decoded".
> >>
> >> I don't know whether this is what one would expect but that should be less of a surprise if documented.
> >>
> >> What do you think?
> >>
> >
> > Yeah, I think it is better to document to avoid any surprises if
> > nobody else sees any problem with it.
>
> Ack.
>

This doesn't seem to be addressed in the latest version. And today, I
think I see one more point about this doc change:
+ <para>
+ A logical replication slot can also be created on a hot standby.
To prevent
+ <command>VACUUM</command> from removing required rows from the system
+ catalogs, <varname>hot_standby_feedback</varname> should be set on the
+ standby. In spite of that, if any required rows get removed, the slot gets
+ invalidated. It's highly recommended to use a physical slot
between the primary
+ and the standby. Otherwise, hot_standby_feedback will work, but
only while the
+ connection is alive (for example a node restart would break it). Existing
+ logical slots on standby also get invalidated if wal_level on
primary is reduced to
+ less than 'logical'.

If hot_standby_feedback is not set then can logical decoding on
standby misbehave? If so, that is not very clear from this doc change
if that is acceptable. One scenario where I think it can misbehave is
if applying WAL records generated after changing wal_level from
'logical' to 'replica' physically removes catalog tuples that could be
referenced by the logical decoding on the standby. Now, as mentioned
in patch 0003's comment in decode.c that it is possible that some
slots may creep even after we invalidate the slots on parameter
change, so while decoding using that slot if some required catalog
tuple has been removed by physical replication then the decoding can
misbehave even before reaching XLOG_PARAMETER_CHANGE record.

--
With Regards,
Amit Kapila.