Horton
Menu

Re: Perform streaming logical transactions by background workers and parallel apply

Amit Kapila <amit.kapila16@gmail.com>

View thread
From: Amit Kapila <amit.kapila16@gmail.com>
To: Dilip Kumar <dilipbalaut@gmail.com>
Cc: "houzj.fnst@fujitsu.com" <houzj.fnst@fujitsu.com>, "wangw.fnst@fujitsu.com" <wangw.fnst@fujitsu.com>, Peter Smith <smithpb2250@gmail.com>, Masahiko Sawada <sawada.mshk@gmail.com>, "shiy.fnst@fujitsu.com" <shiy.fnst@fujitsu.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-08-10T03:38:36Z
Lists: pgsql-hackers 
On Tue, Aug 9, 2022 at 5:39 PM Amit Kapila <amit.kapila16@gmail.com> wrote:
>
> On Tue, Aug 9, 2022 at 11:09 AM Dilip Kumar <dilipbalaut@gmail.com> wrote:
> >
> > Some more comments
> >
> > +    /*
> > +     * Exit if any relation is not in the READY state and if any worker is
> > +     * handling the streaming transaction at the same time. Because for
> > +     * streaming transactions that is being applied in apply background
> > +     * worker, we cannot decide whether to apply the change for a relation
> > +     * that is not in the READY state (see should_apply_changes_for_rel) as we
> > +     * won't know remote_final_lsn by that time.
> > +     */
> > +    if (list_length(ApplyBgworkersFreeList) !=
> > list_length(ApplyBgworkersList) &&
> > +        !AllTablesyncsReady())
> > +    {
> > +        ereport(LOG,
> > +                (errmsg("logical replication apply workers for
> > subscription \"%s\" will restart",
> > +                        MySubscription->name),
> > +                 errdetail("Cannot handle streamed replication
> > transaction by apply "
> > +                           "background workers until all tables are
> > synchronized")));
> > +
> > +        proc_exit(0);
> > +    }
> >
> > How this situation can occur? I mean while starting a background
> > worker itself we can check whether all tables are sync ready or not
> > right?
> >
>
> We are already checking at the start in apply_bgworker_can_start() but
> I think it is required to check at the later point of time as well
> because the new rels can be added to pg_subscription_rel via Alter
> Subscription ... Refresh. I feel if that reasoning is correct then we
> can probably expand comments to make it clear.
>
> > +    /* Check the status of apply background worker if any. */
> > +    apply_bgworker_check_status();
> > +
> >
> > What is the need to checking each worker status on every commit?  I
> > mean if there are a lot of small transactions along with some
> > steamiing transactions
> > then it will affect the apply performance for those small transactions?
> >
>
> I don't think performance will be a concern because this won't do any
> costly operation unless invalidation happens in which case it will
> access system catalogs. However, if my above understanding is correct
> that new tables can be added during the apply process then not sure
> doing it at commit time is sufficient/correct because it can change
> even during the transaction.
>

One idea that may handle it cleanly is to check for
SUBREL_STATE_SYNCDONE state in should_apply_changes_for_rel() and
error out for apply_bg_worker(). For the SUBREL_STATE_READY state, it
should return true and for any other state, it can return false. The
one advantage of this approach could be that the parallel apply worker
will give an error only if the corresponding transaction has performed
any operation on the relation that has reached the SYNCDONE state.
OTOH, checking at each transaction end can also lead to erroring out
of workers even if the parallel apply transaction doesn't perform any
operation on the relation which is not in the READY state.

-- 
With Regards,
Amit Kapila.

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Fix invalid memory access during the shutdown of the parallel apply worker.

  2. Fix assertion failure in apply worker.

  3. Use elog to report unexpected action in handle_streamed_transaction().

  4. Use appropriate wait event when sending data in the apply worker.

  5. Allow the logical_replication_mode to be used on the subscriber.

  6. Rename GUC logical_decoding_mode to logical_replication_mode.

  7. Display the leader apply worker's PID for parallel apply workers.

  8. Improve the code to decide and process the apply action.

  9. Document the newly added wait events added by commit 216a784829.

  10. Perform apply of large transactions by parallel workers.

  11. Wake up a subscription's replication worker processes after DDL.

  12. Add copyright notices to meson files

  13. Better document logical replication parameters

  14. Add a common function to generate the origin name.

  15. Harmonize parameter names in storage and AM code.

  16. Avoid using list_length() to test for empty list.

  17. Improve two comments related to a boolean DefElem's value

  18. Fix partition table's REPLICA IDENTITY checking on the subscriber.

  19. Fix data inconsistency between publisher and subscriber.

  20. Fix cache look-up failures while applying changes in logical replication.