Message-ID: <de7e9c5f-dbaf-48b0-8a65-060615c1e9dd@oss.nttdata.com>
Date: Thu, 20 Mar 2025 02:51:43 +0900
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: pg_recvlogical requires -d but not described on the documentation
Content-Language: en-US
To: "Hayato Kuroda (Fujitsu)" <kuroda.hayato@fujitsu.com>,
 "David G. Johnston" <david.g.johnston@gmail.com>
Cc: Ashutosh Bapat <ashutosh.bapat.oss@gmail.com>,
 "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>,
 'vignesh C' <vignesh21@gmail.com>
References: 
 <OSCPR01MB14966930B4357BAE8C9D68A8AF5C72@OSCPR01MB14966.jpnprd01.prod.outlook.com>
 <CAExHW5uOprZgj2rVQ0HT3ckr+fVvhG-DvcVT+wfmwnwynH00pw@mail.gmail.com>
 <OSCPR01MB149660D1F436F8B82912FCB53F5C72@OSCPR01MB14966.jpnprd01.prod.outlook.com>
 <CAExHW5v94uRcr4f3+yhLrLX8UeYusJB06mcSqfdvu8fYUmVc=Q@mail.gmail.com>
 <OS7PR01MB14968CC4F20B99B8E1508CE81F5C32@OS7PR01MB14968.jpnprd01.prod.outlook.com>
 <CAKFQuwYWVT84GM2OqRx8EqNrfzNM-zbpQ5Y2bA1dPO9jUgo_Kg@mail.gmail.com>
 <b15ecf4f-e5af-4fbb-82c2-a425f453e0b2@oss.nttdata.com>
 <OSCPR01MB14966F6BA7B93EF637D256394F5DE2@OSCPR01MB14966.jpnprd01.prod.outlook.com>
 <OSCPR01MB149668F472C57FEBDCC480E90F5DE2@OSCPR01MB14966.jpnprd01.prod.outlook.com>
 <d235c2a6-fe1e-4bd1-96fa-2e2c884f211d@oss.nttdata.com>
 <OSCPR01MB149660CCE2ADCC54A11F57E09F5D92@OSCPR01MB14966.jpnprd01.prod.outlook.com>
From: Fujii Masao <masao.fujii@oss.nttdata.com>
In-Reply-To: 
 <OSCPR01MB149660CCE2ADCC54A11F57E09F5D92@OSCPR01MB14966.jpnprd01.prod.outlook.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Archived-At: 
 <https://www.postgresql.org/message-id/de7e9c5f-dbaf-48b0-8a65-060615c1e9dd%40oss.nttdata.com>
Precedence: bulk


On 2025/03/19 11:32, Hayato Kuroda (Fujitsu) wrote:
> Dear Fujii-san,
> 
>> It looks like commit 0c013e08cfb introduced a bug that causes "pg_recvlogical
>> --drop-slot"
>> without --dbname to check whether it's connected to a specific database and fail
>> if it's not.
>>
>> This commit was added before 9.5, while pg_recvlogical was introduced in 9.4. On
>> my env,
>> "pg_recvlogical --drop-slot" without --dbname worked as expected in 9.4 but
>> started
>> failing in 9.5 or later.
>>
>> So, I think the proper fix is to avoid raising a fatal error even when not connected
>> to
>> a specific database in --drop-slot action.
> 
> +1. I created patch to fix it. 0001 was completely same as you did.

Thanks for the patch! It looks good to me.

I'm considering whether to back-patch these changes to older versions.
Since pg_recvlogical --drop-slot worked without --dbname in 9.4
but started failing unintentionally in 9.5, it could be considered a bug.
However, this behavior has existed for a long time without complaints or
bug reports, and there was no clear documentation stating that
--drop-slot should work without --dbname.

Given this, I think that also we could treat it as not a bug and apply
the change only to the master branch. What do you think should we
back-patch it as a bug fix or apply it only to master?

Regards,

-- 
Fujii Masao
Advanced Computing Technology Center
Research and Development Headquarters
NTT DATA CORPORATION