Subject: Re: pg_upgrade version checking questions
To: Daniel Gustafsson <daniel@yesql.se>
Cc: "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>
References: <9328.1552952117@sss.pgh.pa.us>
 <pC-NMmh4vQLQP76YTwY4AuoD4OdNw9egikekyQpXFpgqmTlGjIZXOTd2W5RDZPpRski5N3ADRrLYgLk6QUuvmuT5fWC9acPAYyDU1AVxJcU=@yesql.se>
 <20190327124351.GG12804@msg.df7cb.de>
 <L-956-YkM0DEV1jOI3O32iVnD4I4ZMoqfOPwots1c4Q1dI9jNOn4QSlANcQk2FwJE4Yv_sXIuT5Ty759nITA4BxUq_znsxGrVaC5BTP5buU=@yesql.se>
 <29b364fe-b1f9-5171-6f78-1f01a94244ac@2ndquadrant.com>
 <F0DDA232-983B-44C3-98E1-0DD02746DFDF@yesql.se>
 <3177834d-8313-c641-2d20-f99d69d8b6ff@2ndquadrant.com>
 <3CE32AD0-678E-42A8-A2B1-C2A3EDDE15C7@yesql.se>
 <cd1fdac1-6334-b1b3-f74e-9332178ed0f4@2ndquadrant.com>
 <22906594-DDD7-4085-8318-595C0C9C75C6@yesql.se>
 <e048d212-f72e-cff5-36fd-3e867e7d7267@2ndquadrant.com>
 <DFE4CCFB-9A3A-4706-A2B5-A635EC148634@yesql.se>
From: Peter Eisentraut <peter.eisentraut@2ndquadrant.com>
Message-ID: <35ec2a77-2d38-d65e-fe9f-737a0c0c2955@2ndquadrant.com>
Date: Wed, 3 Mar 2021 09:57:38 +0100
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.16; rv:78.0)
 Gecko/20100101 Thunderbird/78.7.1
MIME-Version: 1.0
In-Reply-To: <DFE4CCFB-9A3A-4706-A2B5-A635EC148634@yesql.se>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 02.03.21 22:51, Daniel Gustafsson wrote:
>> The commit message says something about "to ensure the health of the target cluster", which doesn't make sense to me.  Maybe find a better wording.
> 
> Reworded in the attached updated version.
> 
>> The name find_exec() seems not very accurate.  It doesn't find anything.  Maybe "check"?
> 
> I'm not wild about check_exec(), but every other name I could think of was
> drastically worse so I went with check_exec.
> 
>> I'm not sure why the new find_exec() adds EXE.  AFAIK, this is only required for stat(), and validate_exec() already does it.
> 
> Good point, fixed.

I committed this.  I added a pg_strip_crlf() so that there are no 
newlines in the error message.  I also slightly reworded the error 
message to make the found and expected value distinguishable.