Subject: Re: archive status ".ready" files may be created too early
To: Kyotaro Horiguchi <horikyota.ntt@gmail.com>, hlinnaka@iki.fi
Cc: matsumura.ryo@fujitsu.com, bossartn@amazon.com, masao.fujii@gmail.com,
 pgsql-hackers@postgresql.org
References: <20200706.141340.695077532820840879.horikyota.ntt@gmail.com>
 <OSAPR01MB50270BF01719418683625CECE8660@OSAPR01MB5027.jpnprd01.prod.outlook.com>
 <65e309c0-5d51-7f9e-80ed-947414db8280@iki.fi>
 <20201014.090628.839639906081252194.horikyota.ntt@gmail.com>
From: Anastasia Lubennikova <a.lubennikova@postgrespro.ru>
Message-ID: <6009933f-5b21-db6a-a23a-14a414aba94d@postgrespro.ru>
Date: Sun, 29 Nov 2020 22:16:58 +0300
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.10.0
MIME-Version: 1.0
In-Reply-To: <20201014.090628.839639906081252194.horikyota.ntt@gmail.com>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Content-Language: en-US
Precedence: bulk

On 14.10.2020 03:06, Kyotaro Horiguchi wrote:
> The patch includes a fix for primary->standby case. But I'm not sure
> we can do that in the cascaded case. A standby is not aware of the
> structure of a WAL blob and has no idea of up-to-where to send the
> received blobs. However, if we can rely on the behavior of CopyData
> that we always receive a blob as a whole sent from the sender at once,
> the cascaded standbys are free from the trouble (as far as the
> cascaded-standby doesn't crash just before writing the last-half of a
> record into pg_wal and after archiving the last full-segment, which
> seems unlikely.).
>
> regards.
>

Status update for a commitfest entry.

This entry was "Waiting on author" during this CF. As I see, the latest 
message contains new version of the patch.
Does it need more work? Are you going to continue working on it?

-- 
Anastasia Lubennikova
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company