From: Tom Lane <tgl@sss.pgh.pa.us>
To: Jan Wieck <jan@wi3ck.info>
cc: Bruce Momjian <bruce@momjian.us>, Zhihong Yu <zyu@yugabyte.com>,
        Andrew Dunstan <andrew@dunslane.net>,
        Magnus Hagander <magnus@hagander.net>,
        Robins Tharakan <tharakan@gmail.com>,
        Peter Eisentraut <peter.eisentraut@enterprisedb.com>,
        "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Subject: Re: pg_upgrade failing for 200+ million Large Objects
In-reply-to: <6cccaa33-c263-b8a2-b064-985605d33d25@wi3ck.info>
References: <181907.1616253799@sss.pgh.pa.us>
 <147fa478-510b-18ef-5323-9c1725b2493c@wi3ck.info>
 <c2a43a97-e551-ea6d-7a4f-a4709b4e0cbd@dunslane.net>
 <5bdcb010-ecdd-c69a-b441-68002fc38483@wi3ck.info>
 <ee7d96b8-7b0e-bb76-9724-900606efe69a@dunslane.net>
 <CALNJ-vSPnudYcS8YowHeq7VAf7PBRyu0vpsbmb_VvTJgguVyHA@mail.gmail.com>
 <3886649c-c77d-dfd7-08a4-d1606bc71254@wi3ck.info>
 <91ccdb0d-42fd-7413-4e7c-3d6445655d2e@wi3ck.info>
 <20210323145628.GD579@momjian.us>
 <8d8d3961-8e8b-3dbe-f911-6f418c5fb1d3@wi3ck.info>
 <20210323180646.GG579@momjian.us>
 <91b02dc1-f0d9-e50d-849c-18d9a66484fb@wi3ck.info>
 <985941.1616524546@sss.pgh.pa.us>
 <c0f5fde5-5cad-65ed-dd30-cd852301d074@wi3ck.info>
 <986904.1616525964@sss.pgh.pa.us>
 <6cccaa33-c263-b8a2-b064-985605d33d25@wi3ck.info>
Comments: In-reply-to Jan Wieck <jan@wi3ck.info>
	message dated "Tue, 23 Mar 2021 15:22:04 -0400"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <988414.1616528159.1@sss.pgh.pa.us>
Date: Tue, 23 Mar 2021 15:35:59 -0400
Message-ID: <988415.1616528159@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/988415.1616528159%40sss.pgh.pa.us>
Precedence: bulk

Jan Wieck <jan@wi3ck.info> writes:
> The problem here is that pg_upgrade itself is invoking a shell again. It 
> is not assembling an array of arguments to pass into exec*(). I'd be a 
> happy camper if it did the latter. But as things are we'd have to add 
> full shell escapeing for arbitrary strings.

Surely we need that (and have it already) anyway?

I think we've stayed away from exec* because we'd have to write an
emulation for Windows.  Maybe somebody will get fed up and produce
such code, but it's not likely to be the least-effort route to the
goal.

			regards, tom lane