Re: Logical Replication vs. 2PC - Mailing list pgsql-hackers

From Amit Kapila
Subject Re: Logical Replication vs. 2PC
Date
Msg-id CAA4eK1K1EHi1cZ=3bM5Ct1YpTP+PTyPLd-PwVqMOu6o0YLmFkw@mail.gmail.com
Whole thread Raw
In response to Re: Logical Replication vs. 2PC  (Markus Wanner <markus.wanner@enterprisedb.com>)
List pgsql-hackers
On Sun, Mar 21, 2021 at 2:47 PM Markus Wanner
<markus.wanner@enterprisedb.com> wrote:
>
> On 20.03.21 16:14, Amit Kapila wrote:
> > Right, but I guess in our case using user-provided GID will conflict
> > if we use multiple subscriptions on the same node. So, it is better to
> > generate a unique identifier like we are discussing here, something
> > like (origin_id of subscription + xid of the publisher). Do you see
> > any problem with that?
>
> No, quite the opposite: I'm the one advocating the use of xids to
> identify transactions.
>

Okay.

>  See my patch for filter_prepare.
>

I'll think once again from this angle and respond on that thread,
probably one use case could be for the plugins which use xid to
generate GID. In such cases, xid might be required to filter the
transaction.

-- 
With Regards,
Amit Kapila.



pgsql-hackers by date:

Previous
From: Pavel Stehule
Date:
Subject: Re: proposal - psql - use pager for \watch command
Next
From: Amit Kapila
Date:
Subject: Re: replication cleanup code incorrect way to use of HTAB HASH_REMOVE ?