On Mon, Nov 7, 2016 at 10:31 PM, MauMau <maumau307@gmail.com> wrote:
> Yes, I tested both your patch and mine. I used the attached pg_ctl.c.
> It adds -z option which disables SECURITY_SERVICE_RID.
Okay, so you did exactly what I did except that you wrapped with an option...
> I guess you registered the service without specifying the service
> account with -U. Then the service runs as the Local System account,
> whence pgwin32_is_service() returns 1.
Thank you, that's as you said. I was just using the local user account
which is why I did not see the difference. And now I can. I finished
by not using your version of pg_ctl but mine still the result is the
same. Hm, now that we are two folks able to test the difference, I'd
suggest that a committer pops up and pushes the one-liner patch I sent
upthread and back-patches it.
For the refactoring, I guess that we could sort that later on, and I
promise to look at soon. The issue reported on this thread has been
here for 1 year, I am glad that someone finally came up an easy way to
test things.
--
Michael