Re: "make check" in src/test/isolation is unworkable

From: Andrew Dunstan <andrew(at)dunslane(dot)net>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: pgsql-hackers(at)postgreSQL(dot)org
Subject: Re: "make check" in src/test/isolation is unworkable
Date: 2011-05-09 12:59:55
Message-ID: 4DC7E54B.1040006@dunslane.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 05/08/2011 09:54 PM, Andrew Dunstan wrote:
>
>
> On 05/08/2011 07:35 PM, Tom Lane wrote:
>> I believe that the "make check" target in src/test/isolation is
>> fundamentally unportable, as is illustrated by the fact that buildfarm
>> member coypu is currently choking on it. The reason is that the
>> pg_isolation_regress program depends on libpq, and in particular it
>> depends on having an *installed* libpq. Anyplace where it appears to
>> work, it's because you already installed Postgres, or at least libpq.
>
> darn, you're right.
>
>

OK, I have crake running the installation checks:
<http://www.pgbuildfarm.org/cgi-bin/show_stage_log.pl?nm=crake&dt=2011-05-09%2004%3A17%3A01&stg=isolation-check>,
so I have checked in a hot fix for the buildfarm client:
<https://github.com/PGBuildFarm/client-code/commit/c3c20a6457f57efcdcecb83e9c8168791f33f699>

What's a bit annoying is that these tests were checked in without a
vestige of MSVC support, and nobody pinged the usual suspects (i.e.
Magnus and me) to ask for help in providing it, unless my memory is even
worse than usual. We have a bit of work to do to enable that, which I'll
try to get done before pgcon.

cheers

andrew

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Roberto Mello 2011-05-09 13:20:50 Re: [HACKERS] Re: New Canadian nonprofit for trademark, postgresql.org domain, etc.
Previous Message Robert Haas 2011-05-09 11:53:50 Re: superusers are members of all roles?