Re: 回复: Why is XLOG_FPI_FOR_HINT always need backups?

From: Kyotaro Horiguchi <horikyota(dot)ntt(at)gmail(dot)com>
To: masao(dot)fujii(at)oss(dot)nttdata(dot)com
Cc: 757634191(at)qq(dot)com, japinli(at)hotmail(dot)com, pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Re: 回复: Why is XLOG_FPI_FOR_HINT always need backups?
Date: 2021-07-19 01:16:18
Message-ID: 20210719.101618.1726097239880367439.horikyota.ntt@gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

At Sat, 17 Jul 2021 00:14:34 +0900, Fujii Masao <masao(dot)fujii(at)oss(dot)nttdata(dot)com> wrote in
> Thanks for updating the patch! It basically looks good to me.
>
> * Full-page image (FPI) records contain nothing else but a backup
> * block (or multiple backup blocks). Every block reference must
> * include a full-page image - otherwise there would be no point in
> * this record.
>
> The above comment also needs to be updated?

In short, no. In contrast to the third paragraph, the first paragraph
should be thought that it is describing XLOG_FPI. However, actually
it is not super obvious so it's better to make it clearer. Addition to
that, it seems to me (yes, to *me*) somewhat confused between "block
reference", "backup block" and "full-page image". So I'd like to
adjust the paragraph as the following.

> * XLOG_FPI records contain nothing else but one or more block
> * references. Every block reference must include a full-page image
> * regardless of the full_page_writes setting - otherwise there would
> * be no point in this record.

FYI (or, for the record), the first paragraph got to the current shape
by the commit 9155580fd5, where XLOG_FPI was modified to be able to
hold more than one block references.

regards.

--
Kyotaro Horiguchi
NTT Open Source Software Center

Attachment Content-Type Size
v3-0001-Make-FPI_FOR_HINT-follow-standard-FPI-emitting-po.patch text/x-patch 4.2 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message John Naylor 2021-07-19 01:26:47 Re: speed up verifying UTF-8
Previous Message Ranier Vilela 2021-07-19 00:38:30 Re: Remove redundant strlen call in ReplicationSlotValidateName