Re: The reason for loris' intermittent prepared_xacts failures

Tom Lane wrote:

>4. Fix the Windows special case so that we test in C locale not en_US
>locale (is this even possible on that platform?)  Should we be using
>--no-locale in pg_regress ... or wouldn't that help?
>
>
>  
>

--no-locale does work on Windows - I recently added it to the buildfarm 
script's initdb settings to handle an odd Windows issue - I forget 
where, but possibly in the PL installcheck tests which I recently got 
working. How things will work with that setting on a non-English locale 
box I don't know - it might help if we had a Windows buildfarm member 
from one of our colleagues in a non-English speaking country.

FWIW, I just added --no-locale to the initdb args in pg_regress.sh on 
loris and got a clean run of make check.

Regarding the race condition, it certainly sounds like something that 
needs more than a bandaid.

cheers

andrew