Re: Logical Replication - test_decoding - unchanged-toast-datum

From: Abhinav Singh <abhinav(dot)thegame(at)gmail(dot)com>
To: Euler Taveira <euler(at)timbira(dot)com(dot)br>
Cc: pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Logical Replication - test_decoding - unchanged-toast-datum
Date: 2017-09-27 00:46:08
Message-ID: CALm8xzj+124HEEGB6vowmwioCdrb=mUGVx-MONwd4KE+PZhT0w@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hello Euler,

Thank you very for your prompt response.

I am really very sorry that I didn't mention about my exact question. So as
you can see from my previous email, in the replication slot I see the value
'*unchanged-toast-datum*' for the column of dataytpe '*character varying*'.

Based on this, when I query my target PostgreSQL instance after the *UPDATE
*statement is run, the value in this column gets changed to '
*unchanged-toast-datum*'. The steps that I gave is a simple reproduction
and I would request you to please give that a try as well.

Hence, my question is how can workaroud this issue or is there a resolution
for this problem.

Kindly review and please share your comments.

Thanks and Regards,
Abhinav Singh

On Tue, Sep 26, 2017 at 6:25 PM, Euler Taveira <euler(at)timbira(dot)com(dot)br> wrote:

> 2017-09-26 1:11 GMT-03:00 Abhinav Singh <abhinav(dot)thegame(at)gmail(dot)com>:
> > 5. On the target, when I do a select * and see that the column with
> > character varying() datatype has changed to 'unchanged-toast-datum'.
> >
> The column "is_toast" didn't change its value to
> "unchanged-toast-datum". It is just a test_decoding convention that
> means that the value is stored in a TOAST table and it was not
> changed. test_decoding doesn't show TOAST values to avoid performance
> problems and to be brief. Try a SELECT in the table and you will see
> that the value is already there.
>
> You didn't write an explicit question but I believe it was your doubt,
> didn't it?
>
>
> --
> Euler Taveira Timbira -
> http://www.timbira.com.br/
> PostgreSQL: Consultoria, Desenvolvimento, Suporte 24x7 e Treinamento
>

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Amit Langote 2017-09-27 01:22:40 Re: path toward faster partition pruning
Previous Message Amit Langote 2017-09-27 00:39:38 Re: path toward faster partition pruning