Re: INSERT ... ON CONFLICT UPDATE/IGNORE 4.0

Andres Freund <andres@anarazel.de> writes:
> prairiedog, without CCA, failed as well
> http://pgbuildfarm.org/cgi-bin/show_log.pl?nm=prairiedog&dt=2015-05-08%2019%3A55%3A11
> different test, but again directly after index creation. So I hope it's
> indeed the indcheckxmin thing.

Oh, interesting.  That definitely suggests it's about machine speed/timing
not CCA per se (prairiedog being quite slow).  Which would fit with the
indcheckxmin theory.
        regards, tom lane