X-Recipient: archive-cygwin AT delorie DOT com
DomainKey-Signature: a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id
	:list-unsubscribe:list-subscribe:list-archive:list-post
	:list-help:sender:message-id:date:from:mime-version:to:subject
	:references:in-reply-to:content-type; q=dns; s=default; b=QI9kJ5
	ZXgzeVCHE/bxm+gnNU8xToOfmq8ZQV2ERURnsP5iJ7x+D41CK/Dn8rvY8/WWiK6v
	7ahulOG65A7n+DyA9aJH3VNC+hI0T4lILLY977ZWCoGLfINQaVpv6Xrhhqp3ELIu
	nkoXTlo0i1EocAx0GF5hoiZwwxqVRNXJhM148=
DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id
	:list-unsubscribe:list-subscribe:list-archive:list-post
	:list-help:sender:message-id:date:from:mime-version:to:subject
	:references:in-reply-to:content-type; s=default; bh=ncHSJIieluS3
	m6m03YmI90MSnuE=; b=fpWqglSbJGZ6rSeBUSJDX0jrYvXkRfPLaLWgEvt/N79S
	G6ThC+XCq2BsexDCOJedqJ6mcti2dZvsTFBGRPRnfOWDRC3YcZhe5N6zAK13rD3H
	98VTk9ULNlaas3PbTnCoQInKM0ZEYPCzvHcna8v+cVZC88qskHh/jjwRjbCsTV0=
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Id: <cygwin.cygwin.com>
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com
Authentication-Results: sourceware.org; auth=none
X-Virus-Found: No
X-Spam-SWARE-Status: No, score=-2.3 required=5.0 tests=AWL,BAYES_00,RP_MATCHES_RCVD,SPF_HELO_PASS,SPF_PASS autolearn=ham version=3.3.2
X-HELO: mx1.redhat.com
Message-ID: <53E238D2.1030404@redhat.com>
Date: Wed, 06 Aug 2014 08:16:50 -0600
From: Eric Blake <eblake AT redhat DOT com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.7.0
MIME-Version: 1.0
To: cygwin AT cygwin DOT com
Subject: Re: syslog function: Bad file descriptor
References: <53E10628 DOT 3C42FD6F AT boland DOT nl> <20140805164842 DOT GA13601 AT calimero DOT vinschen DOT de> <53E117B7 DOT 4C81186A AT boland DOT nl> <20140805190014 DOT GD13601 AT calimero DOT vinschen DOT de> <53E13FFE DOT E7A33C0D AT boland DOT nl> <20140806091941 DOT GK13601 AT calimero DOT vinschen DOT de> <53E211DD DOT 12F8D5C8 AT boland DOT nl>
In-Reply-To: <53E211DD.12F8D5C8@boland.nl>
OpenPGP: url=http://people.redhat.com/eblake/eblake.gpg
Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="AmLTFsGEwJmBiN0kkBVdxeAIGttVW3Tf9"
X-IsSubscribed: yes

--AmLTFsGEwJmBiN0kkBVdxeAIGttVW3Tf9
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

On 08/06/2014 05:30 AM, D. Boland wrote:

>> Without looking into the sources, I'd assume there's a closelog()
>> call missing prior to the descriptor close orgy.  This closelog()
>> call should fix the problem.
>=20
> It is exactly as you say. I found the close() orgy and put a closelog() p=
rior to it.
> Now it works perfectly without corrupting the aliases file (writable to s=
endmail).
>=20
> I'm asking myself if this closing of 253 file descriptors is a sensible t=
hing to do.

Maybe.  There's two schools of thoughts about preventing unintended
actions due to accidentally leaked fds: 1. any time a parent forks an
unknown child, the parent does an orgy close() between fork() and exec()
so that the child starts life clean. 2. any time a child starts from an
unknown parent, the child starts with an orgy close() to ensure it
starts life clean.  [More recently, there is the addition of O_CLOEXEC
to open() and other APIs to allow atomic FD_CLOEXEC setting, and then
you can avoid the orgy close() in style 1; but that still doesn't stop
style 2 paranoid children]

> What would Sendmail be trying to accomplish there? It comments "Be shure =
we have
> enough file descriptors". And: "in 4.4BSD, the table ([of fd's]) can be h=
uge; impose
> a reasonable limit". Bizarre.

Other programs that I know do an orgy close() on startup: bash, tcsh.
So it's not uncommon.  But if you do it first, it MUST be first - before
you open any fds that must not be accidentally closed.  sendmail is
indeed buggy for trying to use syslog() (which uses an fd) prior to
doing the orgy close.

>=20
> Could it be that incoming e-mail is such a volatile process that previous=
 opened
> file descriptors are not closed quick enough? This feels like a crude hac=
k.

Just because the problem hasn't tripped other platforms does not make it
any less of an upstream sendmail bug.

--=20
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org


--AmLTFsGEwJmBiN0kkBVdxeAIGttVW3Tf9
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
Comment: Public key at http://people.redhat.com/eblake/eblake.gpg

iQEcBAEBCAAGBQJT4jjSAAoJEKeha0olJ0NqbuQIAIJwkxLT26zxfDLnLSZK2gEW
NBHmWpJVACiyXEa8cJH72Z7tNWI/E6R+BDVq35Vf649NWc/21dJpjTDfb7d++QXJ
2eG6BIU06mEsZszCd8EVzEuVhrCiSUboneXvnjf0LIYeU2GJFzBRbfj/zJjVMSie
7O31I9o+vyd1Vh0hZ3b1V+BsMWPK2/Is2IYM7oNWuJkX7X13hb0KK+Kdx0WMhnx2
ayXQE/Qoq0aC9U2mdvLQhWEN6EfKgAv3Ig9OCTVXuy6wRYHsUbQWL6Q8GNRl3Dfo
vyLlrquXGJPhRK/t4HXGdUeplHnHsD8agQ04qqnj41dQduYMf4c4I5LeuDPvOsA=
=QcYZ
-----END PGP SIGNATURE-----

--AmLTFsGEwJmBiN0kkBVdxeAIGttVW3Tf9--