Re: Initial Schema Sync for Logical Replication

> > Yes. Do we have any concrete use case where the subscriber is an older
> > version, in the first place?
> >
>
> As per my understanding, it is mostly due to the reason that it can
> work today. Today, during an off-list discussion with Jonathan on this
> point, he pointed me to a similar incompatibility in MySQL
> replication. See the "SQL incompatibilities" section in doc[1]. Also,
> please note that this applies not only to initial sync but also to
> schema sync during replication. I don't think it would be feasible to
> keep such cross-version compatibility for DDL replication.

I think it's possible to make DDL replication cross-version
compatible, by making the DDL deparser version-aware: the deparsed
JSON blob can have a PG version in it, and the destination server can
process the versioned JSON blob by transforming anything incompatible
according to the original version and its own version.

Regards,
Zane