Alvaro Herrera <alvherre@alvh.no-ip.org> writes:
> Sigh. No, I missed that. I'm going to revert it now, because I see that
> skink failed. I don't see how my patch is to blame since it's a timeout
> and we have no logs for the failure; but maybe my patch did make
> something slower and caused some test to go over the time limit.
skink's been doing that intermittently for some time; I wouldn't
assume this patch is especially to blame.
Given the lack of other failures, maybe you should leave it as-is.
It's not really per policy, but after all this is just a beta ...
regards, tom lane