Re: Object identifier types in logical replication binary mode

From: Emre Hasegeli <emre(at)hasegeli(dot)com>
To: Shlok Kyal <shlok(dot)kyal(dot)oss(at)gmail(dot)com>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Object identifier types in logical replication binary mode
Date: 2024-12-24 15:52:25
Message-ID: CAE2gYzyPTrsdzsVkFwEPOmZx4A8EpYfbeaku9oLperu2m_8Ppw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

> I tested with both 'binary = true' and 'binary = false' option while
> creating a subscription. For me replication is working fine and I am
> not getting any errors in both the cases.
> I have also attached the test script.

I modified your test script to demonstrate the problem.

I created another table before the tested one to make sure oids are
different between the nodes, and used the user-created table in the
test.

I also had to change the subscriber data type to text to get "invalid
byte sequence" error. Still the same script works without the error
with (binary = false).

My analysis on the original post was wrong. The subscriber handles it
as an oid. Though, it's still not clear to me this is a desirable
behaviour for the users, because oid's of the objects differ.

Attachment Content-Type Size
test.sh text/x-sh 1.5 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message David Steele 2024-12-24 16:04:20 Re: Fix logging for invalid recovery timeline
Previous Message Tom Lane 2024-12-24 15:52:08 Re: RFC: Allow EXPLAIN to Output Page Fault Information