Date: Thu, 9 Jul 2020 18:20:04 -0400
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: pgbf@twiska.com, Kyotaro Horiguchi <horikyota.ntt@gmail.com>,
	masao.fujii@oss.nttdata.com, pgsql-hackers@lists.postgresql.org
Subject: Re: min_safe_lsn column in pg_replication_slots view
Message-ID: <20200709222004.GA10331@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <2008607.1594331766@sss.pgh.pa.us>
User-Agent: Mutt/1.10.1 (2018-07-13)
Precedence: bulk

On 2020-Jul-09, Tom Lane wrote:

> and even the most cursory look at the code confirms that there's a
> real bug here.  KeepLogSeg expects *logSegNo to be defined on entry,
> but GetWALAvailability hasn't bothered to initialize oldestSlotSeg.
> It is not clear to me which one is in the wrong; the comment for
> KeepLogSeg isn't particularly clear on this.

Oh, so I introduced the bug when I removed the initialization in this
fix.  That one was using the wrong datatype, but evidently it achieved
the right effect.  And KeepLogSeg is using the wrong datatype Invalid
macro also.

I think we should define InvalidXLogSegNo to be ~((uint64)0) and add a
macro to test for that.

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services