Subject: Re: Minimal logical decoding on standbys
To: Ronan Dunklau <ronan.dunklau@aiven.io>, Ibrar Ahmed
	<ibrar.ahmad@gmail.com>, Amit Khandekar <amitdkhan.pg@gmail.com>,
	<pgsql-hackers@lists.postgresql.org>
CC: Andres Freund <andres@anarazel.de>, <fabriziomello@gmail.com>, Alvaro
 Herrera <alvherre@2ndquadrant.com>, tushar <tushar.ahuja@enterprisedb.com>,
	"[pgdg] Robert Haas" <robertmhaas@gmail.com>, Rahila Syed
	<rahila.syed@2ndquadrant.com>, pgsql-hackers <pgsql-hackers@postgresql.org>
References: 
 <CAFcNs+q9FvSW7gMqM__aeWOoDNqfcQCoB8LiO-X33fvqGO8mkQ@mail.gmail.com>
 <CALtqXTfXsyq-G7x3Ti3bm4KgpepqZA_-ZukzJ5W5vFLjRrhGSg@mail.gmail.com>
 <fa90206a-6bf1-2121-cf0c-15b0734a2731@amazon.com>
 <1655255.DiLJNMXF8A@aivenronan>
From: "Drouvot, Bertrand" <bdrouvot@amazon.com>
Message-ID: <c275aaa7-5c57-eda2-5b70-a6774e9bc13b@amazon.com>
Date: Mon, 2 Aug 2021 17:31:46 +0200
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:78.0)
 Gecko/20100101 Thunderbird/78.12.0
MIME-Version: 1.0
In-Reply-To: <1655255.DiLJNMXF8A@aivenronan>
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit
Content-Language: en-US
Archived-At: 
 <https://www.postgresql.org/message-id/c275aaa7-5c57-eda2-5b70-a6774e9bc13b%40amazon.com>
Precedence: bulk

Hi Ronan,

Thanks for looking at it.

On 8/2/21 1:57 PM, Ronan Dunklau wrote:
> Le mardi 27 juillet 2021, 09:23:48 CEST Drouvot, Bertrand a écrit :
>> Thanks for the warning, rebase done and new v21 version attached.
>>
>> Bertrand
> Hello,
>
> I've taken a look at this patch, and it looks like you adressed every prior
> remark, including the race condition Andres was worried about.

I think there is still 2 points that need to be addressed (see [1])

>
> As for the basics: make check-world and make installcheck-world pass.
>
> I think the beahviour when dropping a database on the primary should be
> documented, and proper procedures for handling it correctly should be
> suggested.
>
> Something along the lines of:
>
> "If a database is dropped on the primary server, the logical replication slot
> on the standby will be dropped as well. This means that you should ensure that
> the client usually connected to this slot has had the opportunity to stream
> the latest changes before the database is dropped."

I am not sure we should highlight this as part of this patch.

I mean, the same would currently occur on a non standby if you drop a 
database that has a replication slot linked to it.

>
> As for the patches themselves, I only have two small comments to make.
>
> In patch 0002, in InvalidateConflictingLogicalReplicationSlots, I don't see the
> need to check for an InvalidOid since we already check the SlotIsLogical:
>
> +               /* We are only dealing with *logical* slot conflicts. */
> +               if (!SlotIsLogical(s))
> +                       continue;
> +
> +               /* not our database and we don't want all the database,
> skip */
> +               if ((s->data.database != InvalidOid && s->data.database
> != dboid) && TransactionIdIsValid(xid))
> +                       continue;

Agree, v22 attached in [1] does remove the useless s->data.database != 
InvalidOid check, thanks!

>
> In patch 0004, small typo in the test file:
> +##################################################
> +# Test standby promotion and logical decoding bheavior
> +# after the standby gets promoted.
> +##################################################

Typo also fixed in v22, thanks!

Bertrand

[1]: 
https://www.postgresql.org/message-id/69aad0bf-697a-04e1-df6c-0920ec8fa528%40amazon.com