Horton

Re: Skipping logical replication transactions on subscriber side

Amit Kapila <amit.kapila16@gmail.com>

View thread View on postgr.es
From: Amit Kapila <amit.kapila16@gmail.com>
To: "osumi.takamichi@fujitsu.com" <osumi.takamichi@fujitsu.com>
Cc: Masahiko Sawada <sawada.mshk@gmail.com>, "houzj.fnst@fujitsu.com" <houzj.fnst@fujitsu.com>, "tanghy.fnst@fujitsu.com" <tanghy.fnst@fujitsu.com>, Alexey Lesovsky <lesovsky@gmail.com>, Peter Eisentraut <peter.eisentraut@enterprisedb.com>, PostgreSQL-development <pgsql-hackers@postgresql.org>, Greg Nancarrow <gregn4422@gmail.com>
Date: 2021-10-18T08:48:46Z
Lists: pgsql-hackers 
On Mon, Oct 11, 2021 at 12:57 PM osumi.takamichi@fujitsu.com
<osumi.takamichi@fujitsu.com> wrote:
>
> On Monday, October 11, 2021 11:51 AM Masahiko Sawada <sawada.mshk@gmail.com> wrote:
> > On Fri, Oct 8, 2021 at 4:09 PM osumi.takamichi@fujitsu.com
> > <osumi.takamichi@fujitsu.com> wrote:
> > >
> > > On Thursday, September 30, 2021 2:45 PM Masahiko Sawada
> > <sawada.mshk@gmail.com> wrote:
> > > > I've attached updated patches that incorporate all comments I got so
> > > > far. Please review them.
> > > Sorry, if I misunderstand something but did someone check what happens
> > > when we execute ALTER SUBSCRIPTION ... RESET (streaming) in the middle
> > > of one txn which has several streaming of data to the sub, especially
> > > after some part of txn has been already streamed.
> > > My intention of this is something like *if* we can find an actual harm
> > > of this, I wanted to suggest the necessity of a safeguard or some measure
> > into the patch.
> ...
> > > I observed that the subscriber doesn't accept STREAM_COMMIT in this
> > > case but gets BEGIN&COMMIT instead at the end.
> > > I couldn't find any apparent and immediate issue from those steps but
> > > is that no problem ?
> > > Probably, this kind of situation applies to other reset target options ?
> >
> > I think that if a subscription parameter such as ‘streaming’ and ‘binary’ is
> > changed, an apply worker exits and the launcher starts a new worker (see
> > maybe_reread_subscription()). So I guess that in this case, the apply worker
> > exited during receiving streamed changes, restarted, and received the same
> > changes with ‘streaming = off’, therefore it got BEGIN and COMMIT instead. I
> > think that this happens even by using ‘SET (‘streaming’ = off)’.
> You are right. Yes, I checked that the apply worker did exit
> and the new apply worker process dealt with the INSERT in the above case.
> Also, setting streaming = false was same.
>

I think you can additionally verify that temporary streaming files get
removed after restart.

-- 
With Regards,
Amit Kapila.

Commits

  1. Test ALIGNOF_DOUBLE==4 compatibility under ALIGNOF_DOUBLE==8.

  2. Reorder subskiplsn in pg_subscription to avoid alignment issues.

  3. Add ALTER SUBSCRIPTION ... SKIP.

  4. Optionally disable subscriptions on error.

  5. Update docs of logical replication for commit 8d74fc96db.

  6. Respect permissions within logical replication.

  7. Fix regression test failure caused by commit 8d74fc96db.

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

  9. Add logical change details to logical replication worker errcontext.

  10. Rename LOGICAL_REP_MSG_STREAM_END to LOGICAL_REP_MSG_STREAM_STOP.

  11. Fix typo in protocol.sgml.

  12. Remove unused argument in apply_handle_commit_internal().

  13. Fix replication of in-progress transactions in tablesync worker.

  14. Reorder pg_sequence columns to avoid alignment issue