Date: Fri, 29 Jun 2018 18:50:55 -0400
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: pgsql-hackers@lists.postgresql.org, 2q-andrew@alvin.alvh.no-ip.org
Subject: Re: pgsql: Fix "base" snapshot handling in logical decoding
Message-ID: <20180629225055.cy3e4ecayibwr5rd@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20180629222906.w5hqjgclmtk76v5i@alvherre.pgsql>
User-Agent: NeoMutt/20170306-137-4415bd-dirty (1.8.0)
Precedence: bulk

On 2018-Jun-29, Alvaro Herrera wrote:

> On 2018-Jun-28, Tom Lane wrote:
> 
> > Alvaro Herrera <alvherre@alvh.no-ip.org> writes:
> > > Fix "base" snapshot handling in logical decoding
> > 
> > According to buildfarm member friarbird, and as confirmed here,
> > the contrib/test_decoding/specs/oldest_xmin.spec test added by this
> > commit fails under CLOBBER_CACHE_ALWAYS.
> 
> Hm.  Running this test under CLOBBER_CACHE_ALWAYS I see that the VACUUM
> FULL step takes 31 seconds (and the test succeeds).  This is not
> top-of-the-line hardware by a long shot (Intel(R) Core(TM) i7-4600U CPU
> @ 2.10GHz) but I can believe that the other machines are slower or
> busier.

It does fail in the indicated way in CLOBBER_CACHE_RECURSIVELY, but I
guess that's expected.  It also fails if I reduce the timeout from 60
seconds to 25.

I suppose 60 seconds (isolationtester's default timeout) is just not
enough time for those machines.  We could increase it to 180 seconds and
see if that's enough to make them pass ...

Another possibility is to examine column 'state' in isolationtester.  I
thought we only used that timeout for 'waiting' queries, but I see it's
not doing that.

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