| From: | Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> |
|---|---|
| To: | Andres Freund <andres(at)anarazel(dot)de> |
| Cc: | Dilip Kumar <dilipbalaut(at)gmail(dot)com>, Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>, "shiy(dot)fnst(at)fujitsu(dot)com" <shiy(dot)fnst(at)fujitsu(dot)com>, Kyotaro Horiguchi <horikyota(dot)ntt(at)gmail(dot)com>, "smithpb2250(at)gmail(dot)com" <smithpb2250(at)gmail(dot)com>, "Hayato Kuroda (Fujitsu)" <kuroda(dot)hayato(at)fujitsu(dot)com>, "pgsql-hackers(at)lists(dot)postgresql(dot)org" <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: Force streaming every change in logical decoding |
| Date: | 2022-12-28 02:19:56 |
| Message-ID: | CAA4eK1Kadr6HWDOphwzCxngjHe0np+WOUoewCbpNoz1h2VeQYw@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Wed, Dec 28, 2022 at 1:42 AM Andres Freund <andres(at)anarazel(dot)de> wrote:
>
> On 2022-12-26 14:04:28 +0530, Amit Kapila wrote:
> > Pushed.
>
> I did not follow this thread but saw the commit. Could you explain why a GUC
> is the right mechanism here? The commit message didn't explain why a GUC was
> chosen.
>
> To me an option like this should be passed in when decoding rather than a GUC.
>
For that, we need to have a subscription option for this as we need to
reduce test timing for streaming TAP tests (by streaming, I mean
replication of large in-progress transactions) as well. Currently,
there is no subscription option that is merely used for
testing/debugging purpose and there doesn't seem to be a use for this
for users. So, we didn't want to expose it as a user option. There is
also a risk that if users use it they will face a slowdown.
--
With Regards,
Amit Kapila.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Hayato Kuroda (Fujitsu) | 2022-12-28 02:47:56 | RE: Exit walsender before confirming remote flush in logical replication |
| Previous Message | Justin Pryzby | 2022-12-28 01:06:55 | Re: Allow placeholders in ALTER ROLE w/o superuser |