From owner-freebsd-audit Tue Dec 12 18:12: 7 2000 From owner-freebsd-audit@FreeBSD.ORG Tue Dec 12 18:12:05 2000 Return-Path: Delivered-To: freebsd-audit@freebsd.org Received: from citusc.usc.edu (citusc.usc.edu [128.125.38.123]) by hub.freebsd.org (Postfix) with ESMTP id 0D7B637B402; Tue, 12 Dec 2000 18:12:05 -0800 (PST) Received: (from kris@localhost) by citusc.usc.edu (8.9.3/8.9.3) id SAA11088; Tue, 12 Dec 2000 18:13:22 -0800 Date: Tue, 12 Dec 2000 18:13:22 -0800 From: Kris Kennaway To: Kris Kennaway Cc: Warner Losh , Will Andrews , Chris Faulhaber , freebsd-audit@FreeBSD.ORG Subject: Re: mktemp(1) usage Message-ID: <20001212181322.C10901@citusc.usc.edu> References: <20001209171334.J671@puck.firepipe.net> <20001209150853.A57045@peitho.fxp.org> <20001209171334.J671@puck.firepipe.net> <200012100529.WAA26442@harmony.village.org> <20001212180805.B10901@citusc.usc.edu> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="aVD9QWMuhilNxW9f" Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: <20001212180805.B10901@citusc.usc.edu>; from kris@FreeBSD.ORG on Tue, Dec 12, 2000 at 06:08:05PM -0800 Sender: kris@citusc.usc.edu Sender: owner-freebsd-audit@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG --aVD9QWMuhilNxW9f Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Dec 12, 2000 at 06:08:05PM -0800, Kris Kennaway wrote: > On Sat, Dec 09, 2000 at 10:29:50PM -0700, Warner Losh wrote: > > In message <20001209171334.J671@puck.firepipe.net> Will Andrews writes: > > : On Sat, Dec 09, 2000 at 03:08:54PM -0500, Chris Faulhaber wrote: > > : > Would it be more appropriate for scripts such as periodic(8) to > > : > call mktemp(1) using the -t flag. In addition to using TMPDIR, > > : > this allows the use of the system's _PATH_TMP instead of > > : > hardcoding /tmp. > > :=20 > > : Maybe that method doesn't use a random enough number to avoid file > > : races? > >=20 > > If it doesn't, then maybe it should, don't you think. But I think it > > does. We're looking at 8 X's. It would also be a good place to bump > > it to 10 if we needed to, say. >=20 > No need - with the old method of mktemp() encoding you needed more > than 6 X's to be secure, but with the new dense encoding even 6 is > fine. Moreover, mktemp(1) is actually mkstemp(1), so there was never > any problem with it unless you use -u. Forgot to mention that the new mktemp() hasn't yet been MFC'ed - it's in my merge folder but I didn't get time to do it before I left the US. That should be done before merging any patches which reduce the number of X's in use. Kris --aVD9QWMuhilNxW9f Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.4 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE6NttCWry0BWjoQKURAhV+AJ4g1V1tsbPbEwnbKhJSgdTtMu9GNwCglMcv IYuGr9qzG9t7+l46SHLCrCQ= =MZDr -----END PGP SIGNATURE----- --aVD9QWMuhilNxW9f-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-audit" in the body of the message