Content-Type: text/plain; charset=UTF-8
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Luke Koops <luke.koops@entrust.com>,
	Pg Bugs <pgsql-bugs@postgresql.org>
Subject: Re: BUG #4958: Stats collector hung on WaitForMultipleObjectsEx while
	attempting to recv a datagram
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: Robert Haas <robertmhaas@gmail.com>
In-reply-to: 
 <CA+TgmobgZtEobmee7t9ozF3L5Rrn6W1_qX4w5FCFLtkjRNwbFA@mail.gmail.com>
References: <200907311459.n6VExuSa036743@wwwmaster.postgresql.org>
	<CA+TgmoZheGK5AvR8Nw0WPwwxzvfpzFENdCbP_2ennJSBnraEnA@mail.gmail.com>
	<1410.1344363743@sss.pgh.pa.us>
	<CA+TgmobQ0sLjzHdEQZr9QG7n2r-BKej+UTH7rcPPh09doMvUDg@mail.gmail.com>
	<1345576064-sup-7145@alvh.no-ip.org>
	<9323.1345581034@sss.pgh.pa.us>
	<CA+TgmoanozBPOBL0qB4OzP4Jn3xJHYaunwwfmRrT1sBZhnSgew@mail.gmail.com>
	<28583.1345688651@sss.pgh.pa.us>
	<CA+TgmobgZtEobmee7t9ozF3L5Rrn6W1_qX4w5FCFLtkjRNwbFA@mail.gmail.com>
Date: Sat, 25 Aug 2012 00:55:22 -0400
Message-Id: <1345870202-sup-3003@alvh.no-ip.org>
User-Agent: Sup/git
Content-Transfer-Encoding: quoted-printable

Excerpts from Robert Haas's message of s=C3=A1b ago 25 00:10:57 -0400 201=
2:

> > It looks sane to me in a quick once-over (bearing in mind I can't tes=
t
> > it).  What's bothering you about the ereport calls?
>=20
> Well, 9.1 currently has:
>=20
> errmsg_internal("failed to enumerate network events: %i", (int) GetLast=
Error()))
>=20
> Prior to your patch, master had:
>=20
> errmsg_internal("failed to enumerate network events: error code %lu",
> GetLastError())
>=20
> And you changed it to:
>=20
> elog(ERROR, "failed to enumerate network events: error code %u",
> WSAGetLastError())
>=20
> So obviously there's a conflict there.  In the previously-posted
> version, I chose to just take your version, but I'm not sure whether
> it would be better -- in the back branches -- to simply replace
> GetLastError() with WSAGetLastError() and leave it alone otherwise.

Note that for translations none of this matters, because
errmsg_internal() does not mark its argument for translation, and
neither does elog().

The change of %i to "error code %lu" was done recently (during 9.2 devel
cycle) to make all error messages involving GetLastError() consistent;
see commit 1af55e275.

--=20
=C3=81lvaro Herrera                http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services