Re: [BUG]Update Toast data failure in logical replication

Поиск
Список
Период
Сортировка
От Dilip Kumar
Тема Re: [BUG]Update Toast data failure in logical replication
Дата
Msg-id CAFiTN-uS0UhHN0juRjjWA1aDMybnvNfqNtVbSQdap_1g6W_nGQ@mail.gmail.com
обсуждение исходный текст
Ответ на RE: [BUG]Update Toast data failure in logical replication  ("tanghy.fnst@fujitsu.com" <tanghy.fnst@fujitsu.com>)
Ответы RE: [BUG]Update Toast data failure in logical replication  ("tanghy.fnst@fujitsu.com" <tanghy.fnst@fujitsu.com>)
Список pgsql-hackers
Дерево обсуждения
On Mon, 31 May 2021 at 3:33 PM, tanghy.fnst@fujitsu.com <tanghy.fnst@fujitsu.com> wrote:
On Mon, May 31, 2021 5:12 PM Dilip Kumar <dilipbalaut@gmail.com> wrote:
>
> The problem is if the key attribute is not changed we don't log it as
> it should get logged along with the updated tuple, but if it is
> externally stored then the complete key will never be logged because
> there is no log from the toast table.  For fixing this if the key is
> externally stored then always log that.
>
> Please test with the attached patch.

Thanks for your patch. I tested it and the bug was fixed.

Thanks for confirming this.
 
I'm still trying to understand your fix, please allow me to ask more(maybe silly) questions if I found any.

+        * if the key hasn't changedand we're only logging the key, we're done.

I think "changedand" should be "changed and".

Okay, I will fix it.  Lets see what others have to say about this fix, if we agree with this then I think we might have to change the test output. I will do that in the next version along with your comment fix.

В списке pgsql-hackers по дате отправления:

Предыдущее
От: "tanghy.fnst@fujitsu.com"
Дата:
Сообщение: RE: [BUG]Update Toast data failure in logical replication
Следующее
От: Dilip Kumar
Дата:
Сообщение: Re: Decoding speculative insert with toast leaks memory