Re: row filtering for logical replication

Amit Kapila <amit.kapila16@gmail.com>

From: Amit Kapila <amit.kapila16@gmail.com>
To: Greg Nancarrow <gregn4422@gmail.com>
Cc: Peter Smith <smithpb2250@gmail.com>, Ajin Cherian <itsajin@gmail.com>, Euler Taveira <euler@eulerto.com>, "houzj.fnst@fujitsu.com" <houzj.fnst@fujitsu.com>, vignesh C <vignesh21@gmail.com>, "tanghy.fnst@fujitsu.com" <tanghy.fnst@fujitsu.com>, Dilip Kumar <dilipbalaut@gmail.com>, Rahila Syed <rahilasyed90@gmail.com>, Peter Eisentraut <peter.eisentraut@enterprisedb.com>, Önder Kalacı <onderkalaci@gmail.com>, japin <japinli@hotmail.com>, Michael Paquier <michael@paquier.xyz>, David Steele <david@pgmasters.net>, Craig Ringer <craig@2ndquadrant.com>, Amit Langote <amitlangote09@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2021-12-15T10:13:21Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →
  1. Release cache tuple when no longer needed

  2. Add some additional tests for row filters in logical replication.

  3. Fix one of the tests introduced in commit 52e4f0cd47.

  4. Allow specifying row filters for logical replication of tables.

  5. Move scanint8() to numutils.c

  6. Replace Test::More plans with done_testing

  7. Reduce relcache access in WAL sender streaming logical changes

  8. Small cleanups related to PUBLICATION framework code

  9. Add a view to show the stats of subscription workers.

  10. Allow publishing the tables of schema.

  11. Doc: improve documentation of CREATE/ALTER SUBSCRIPTION.

  12. Add PublicationTable and PublicationRelInfo structs

  13. Remove unused argument "txn" in maybe_send_schema().

  14. Add prepare API support for streaming transactions in logical replication.

  15. Unify PostgresNode's new() and get_new_node() methods

  16. Use l*_node() family of functions where appropriate

  17. Add support for prepared transactions to built-in logical replication.

  18. Restore the portal-level snapshot after procedure COMMIT/ROLLBACK.

  19. Rename a parse node to be more general

  20. Remove unused column atttypmod from initial tablesync query

  21. SEARCH and CYCLE clauses

On Wed, Dec 15, 2021 at 1:52 PM Greg Nancarrow <gregn4422@gmail.com> wrote:
>
> On Wed, Dec 15, 2021 at 5:25 PM Amit Kapila <amit.kapila16@gmail.com> wrote:
> >
> > > "If a subscriber is a pre-15 version, the initial table
> > > synchronization won't use row filters even if they are defined in the
> > > publisher."
> > >
> > > Won't this lead to data inconsistencies or errors that otherwise
> > > wouldn't happen?
> > >
> >
> > How? The subscribers will get all the initial data.
> >
>
> But couldn't getting all the initial data (i.e. not filtering) break
> the rules used by the old/new row processing (see v46-0003 patch)?
> Those rules effectively assume rows have been previously published
> with filtering.
> So, for example, for the following case for UPDATE:
>     old-row (no match)    new row (match)  -> INSERT
> the old-row check (no match) infers that the old row was never
> published, but that row could in fact have been in the initial
> unfiltered rows, so in that case an INSERT gets erroneously published
> instead of an UPDATE, doesn't it?
>

But this can happen even when both the publisher and subscriber are
from v15, say if the user defines filter at some later point or change
the filter conditions by Alter Publication. So, not sure if we need to
invent something new for this.

> > > Should such subscriptions be allowed?
> > >
> >
> > I am not sure what you have in mind here? How can we change the
> > already released code pre-15 for this new feature?
> >
>
> I was thinking such subscription requests could be rejected by the
> server, based on the subscriber version and whether the publications
> use filtering etc.
>

Normally, the client sends some parameters to the server like
(streaming, two_pc, etc.) based on which server can take such
decisions. We may need to include some such thing which I am not sure
is required for this particular case especially because that can
happen otherwise as well.

-- 
With Regards,
Amit Kapila.