Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 28 Jul 2011 19:31:14 -0400
From:      Jason Hellenthal <jhell@DataIX.net>
To:        Robert Watson <rwatson@freebsd.org>
Cc:        Glen Barber <gjb@freebsd.org>, svn-src-all@freebsd.org, src-committers@freebsd.org, svn-src-stable-8@freebsd.org, svn-src-stable@freebsd.org
Subject:   Re: svn commit: r224462 - stable/8/usr.sbin/jail
Message-ID:  <20110728233114.GA37774@DataIX.net>
In-Reply-To: <alpine.BSF.2.00.1107281039110.30580@fledge.watson.org>
References:  <201107270156.p6R1uquD035835@svn.freebsd.org> <20110728021914.GA55550@DataIX.net> <4E30CEEB.107@FreeBSD.org> <alpine.BSF.2.00.1107281039110.30580@fledge.watson.org>

next in thread | previous in thread | raw e-mail | index | archive | help

--a8Wt8u1KmwUX3Y2C
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable



On Thu, Jul 28, 2011 at 10:40:19AM +0100, Robert Watson wrote:
>=20
> On Wed, 27 Jul 2011, Glen Barber wrote:
>=20
> >> How is either one of these different ?
> >>
> >> All mv(1) is doing is a cp(1) & rm(1). In either case the filehandle i=
s=20
> >> still broken and a process is not going to just get up and move with i=
t. On=20
> >> the other side though if you copied a pipe or socket or something simi=
liar=20
> >> for example into a jail then it might make whatever is outside availab=
le to=20
> >> the jailed environment.
> >>
> >> Is there something I am misunderstanding about this ? has the way cp(1=
),=20
> >> rm(1) & mv(1) been changed recently ? or is this wording a little off ?
> >
> > The text in the example is just an example of a situation where it may =
be=20
> > possible for a process within a jail(8) to gain filesystem access outsi=
de of=20
> > the jail(8).
>=20
> I wonder, if on these grounds, we should actually advise administrators t=
hat=20
> it is a more robust configuration, both in terms of managing free space a=
nd=20
> avoiding potential escape paths, to put each jail in its own file system.=
=20
> Lots of people do this anyway, and as recommendations go, it's not a bad =
one.=20
> We can then caution that if you *don't* do this, then you need to be care=
ful=20
> about the mv issue.
>=20

That sounds like a perfectly sane idea.

--a8Wt8u1KmwUX3Y2C
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.17 (FreeBSD)
Comment: http://bit.ly/0x89D8547E

iQEcBAEBAgAGBQJOMfFCAAoJEJBXh4mJ2FR+XXkH/im3tgkucIQrtuqd5DZbBIck
q3qtWsT3fGyRoEKB3sGXLzxDpMeic9Rm5qES9FQFM/vW/sxoNu6k8jd/7LwoD5bG
merDyuJCSuHIoaYDGssKUCt/z+sr5+7dwPQEhWsc2MdoDaX55JLiyDsdevbFZoeb
1q0XQVh6jUBHHVzam7ex4xSCZ3UjVwe8gaHbZd3J7uvXPowgHVDcgYbRMDKvQ9Km
xgoWSpXfGR4SCsbzLKVNI4eVN2TvDo80DLqCDFkKr9rOqDB/l2IveZMgq5q8dTZ/
lBkAZEIZZHbANwsfFchGVbfOn6WoKSdjDyiDehwiKTNZYUnLUwJ5tUQI0cdgGqI=
=OQAQ
-----END PGP SIGNATURE-----

--a8Wt8u1KmwUX3Y2C--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20110728233114.GA37774>