From mboxrd@z Thu Jan 1 00:00:00 1970 From: ebiederm@xmission.com (Eric W. Biederman) Subject: Re: [PATCH] [RFC] mnt: add ability to clone mntns starting with the current root Date: Tue, 07 Oct 2014 14:42:18 -0700 Message-ID: <87zjd7r1z9.fsf@x220.int.ebiederm.org> References: <1412683977-29543-1-git-send-email-avagin@openvz.org> <20141007133039.GG7996@ZenIV.linux.org.uk> <20141007133339.GH7996@ZenIV.linux.org.uk> <87r3yjy64e.fsf@x220.int.ebiederm.org> <20141007204627.GI28519@ubuntumail> <87wq8bvbzg.fsf@x220.int.ebiederm.org> <20141007213257.GJ28519@ubuntumail> Mime-Version: 1.0 Content-Type: text/plain Return-path: In-Reply-To: <20141007213257.GJ28519@ubuntumail> (Serge Hallyn's message of "Tue, 7 Oct 2014 21:32:57 +0000") Sender: linux-fsdevel-owner@vger.kernel.org To: Serge Hallyn Cc: Al Viro , Andrey Vagin , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-api@vger.kernel.org, Andrey Vagin , Andrew Morton , Cyrill Gorcunov , Pavel Emelyanov , Serge Hallyn , Rob Landley List-Id: linux-api@vger.kernel.org Serge Hallyn writes: > Quoting Eric W. Biederman (ebiederm@xmission.com): >> What I meant is that it isn't about containers. It is about something >> root can do. So this is not a "container" problem. > > Oh, ok. > > Sorry, I'm getting the two thread confused anyway. I'm going to bow out > here until I can pay proper attention. I think there is half a point here that may be legit, when you are using mount namespaces to jail applications, there may be a problem with umounting / and making it to the underlying rootfs filesystem. I am squinting and looking this way and that but while I can imagine someone more clever than I can think up some unique property of rootfs that makes it a little more exploitable than just mounting a ramfs, but since you have to be root to exploit those properties I think the game is pretty much lost. >> >> So it is only root (and not root in a container) who can get to the >> >> exposed rootfs. >> >> >> >> I have a vague memory someone actually had a real use in miminal systems >> >> for being able to get back to the rootfs and being able to use rootfs as >> >> the rootfs. There was even a patch at that time that Andrew Morton was >> >> carrying for a time to allow unmounting root and get at rootfs, and to >> >> prevent the oops on rootfs unmount in some way. >> >> >> >> So not only do I not think it is a bug to get back too rootfs, I think >> >> it is a feature that some people have expressed at least half-way sane >> >> uses for. >> > >> > They can still do that if they want, using chroot :) >> >> It would take fchdir or fchroot and a directory file descriptor open on >> rootfs. Frequently there is no appropriate directory file descriptor. > > ? you can always escape if you're simply chrooted. waterbuffalo :) filesystem type rootfs. Eric