On 11/30/13, 6:57 PM, Tom Lane wrote:
> Oliver Seemann <oseemann@gmail.com> writes:
>> And it does so, most of of the time. But when run repeatedly in a loop like
>> in the attached script, then it will occasionally return 2 rows with two
>> different id values, something the LIMIT 1 should prevent. In my tests it
>> took from anywhere between 0 to 10 minutes and on average 1 to 2 minutes to
>> trigger the problem.
>
> I failed to reproduce the claimed misbehavior in git tip of any active
> branch. I'd like to think this means we fixed the problem in the last
> two months, but I don't see anything that looks like a promising candidate
> in the commit logs. Perhaps there is some important contributing factor
> you've not mentioned --- nondefault postgresql.conf settings, for
> instance.
I've managed to reproduce this against REL9_1_STABLE
(4bdccd8427718f9c468e5e03286252f37ea771b5). I'm on OS X mavericks,
configure line: ./configure --enable-debug --with-openssl --with-perl
--with-libxml --with-readline. Compiler Apple LLVM version 5.0
(clang-500.2.79) (based on LLVM 3.3svn). No changes to postgresql.conf.
Does that help? Do you need some more information?
Regards,
Marko Tiikkaja
