| From: | Ajin Cherian <itsajin(at)gmail(dot)com> |
|---|---|
| To: | Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> |
| Cc: | PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>, Peter Smith <smithpb2250(at)gmail(dot)com>, Markus Wanner <markus(dot)wanner(at)enterprisedb(dot)com>, osumi(dot)takamichi(at)fujitsu(dot)com |
| Subject: | Re: Logical Replication vs. 2PC |
| Date: | 2021-03-19 05:50:59 |
| Message-ID: | CAFPTHDYeXL7aHWUW41drP7zWvOE4OdbzSzqcVK-dy8c3DELsPA@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Thu, Mar 18, 2021 at 8:46 PM Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
>
>
> However, if the user has setup synchronous_standby_names for all the
> subscriptions then we won't be able to proceed because the prepare on
> publisher will wait for all the subscriptions to ack and the
> subscriptions are waiting for the first prepare to finish.
>
But is it a valid use case to have two synchronous standbys which are two
subscriptions that are on the same server both with 2pc enabled?
If the purpose of synchronous standby is for durability to prevent data
loss, then why split your tables across 2 subscriptions which are on the
same server?
Maybe it could be documented warning users from having such a setup. Do we
really want to create a solution for an impractical scenario?
regards,
Ajin Cherian
Fujitsu Australia
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Shinya11.Kato | 2021-03-19 06:06:47 | RE: Wrong statistics for size of XLOG_SWITCH during pg_waldump. |
| Previous Message | Bharath Rupireddy | 2021-03-19 05:32:45 | Re: Parallel Inserts in CREATE TABLE AS |