From: Simo Sorce <ssorce-H+wXaHxf7aLQT0dZR+AlfA@public•gmane.org>
To: Andy Lutomirski <luto-kltTT9wpgjJwATOyAt5JVQ@public•gmane.org>
Cc: Vivek Goyal <vgoyal-H+wXaHxf7aLQT0dZR+AlfA@public•gmane.org>,
Daniel J Walsh <dwalsh-H+wXaHxf7aLQT0dZR+AlfA@public•gmane.org>,
David Miller <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public•gmane.org>,
Tejun Heo <tj-DgEjT+Ai2ygdnm+yROfE0A@public•gmane.org>,
"linux-kernel-u79uwXL29TY76Z2rM5mHXA@public•gmane.org"
<linux-kernel-u79uwXL29TY76Z2rM5mHXA@public•gmane.org>,
lpoetter-H+wXaHxf7aLQT0dZR+AlfA@public•gmane.org,
cgroups-u79uwXL29TY76Z2rM5mHXA@public•gmane.org,
kay-H+wXaHxf7aLQT0dZR+AlfA@public•gmane.org,
Network Development
<netdev-u79uwXL29TY76Z2rM5mHXA@public•gmane.org>
Subject: Re: [PATCH 2/2] net: Implement SO_PASSCGROUP to enable passing cgroup path
Date: Thu, 17 Apr 2014 14:31:24 -0400 [thread overview]
Message-ID: <1397759484.2628.93.camel@willson.li.ssimo.org> (raw)
In-Reply-To: <CALCETrUon7mZzp12th2bZ=cJyTjn8ePwg_VtPWL_bykjnnpKLw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
On Thu, 2014-04-17 at 11:04 -0700, Andy Lutomirski wrote:
> On Thu, Apr 17, 2014 at 10:52 AM, Simo Sorce <ssorce-H+wXaHxf7aLQT0dZR+AlfA@public•gmane.org> wrote:
> > On Thu, 2014-04-17 at 10:26 -0700, Andy Lutomirski wrote:
> >> On Thu, Apr 17, 2014 at 10:12 AM, Vivek Goyal <vgoyal-H+wXaHxf7aLQT0dZR+AlfA@public•gmane.org> wrote:
> >> > On Thu, Apr 17, 2014 at 09:55:08AM -0700, Andy Lutomirski wrote:
> >> >> On Thu, Apr 17, 2014 at 9:48 AM, Simo Sorce <ssorce-H+wXaHxf7aLQT0dZR+AlfA@public•gmane.org> wrote:
> >> >> > On Thu, 2014-04-17 at 09:37 -0700, Andy Lutomirski wrote:
> >> >> >> On Thu, Apr 17, 2014 at 9:24 AM, Simo Sorce <ssorce-H+wXaHxf7aLQT0dZR+AlfA@public•gmane.org> wrote:
> >> >> >> > On Thu, 2014-04-17 at 09:11 -0700, Andy Lutomirski wrote:
> >> >> >> >>
> >> >> >> >> No. The logging daemon thinks it wants to know who the writer is, but
> >> >> >> >> the logging daemon is wrong. It actually wants to know who composed a
> >> >> >> >> log message destined to it. The caller of write(2) may or may not be
> >> >> >> >> the same entity.
> >> >> >> >
> >> >> >> > This works both ways, and doesn't really matter, you are *no* better off
> >> >> >> > w/o this interface.
> >> >> >> >
> >> >> >> >> If this form of SO_PASSCGROUP somehow makes it into a pull request for
> >> >> >> >> Linus, I will ask him not to pull it and/or to revert it. I think
> >> >> >> >> he'll agree that write(2) MUST NOT care who called it.
> >> >> >> >
> >> >> >> > And write() does not, there is no access control check being performed
> >> >> >> > here. This call is the same as getting the pid of the process and
> >> >> >> > crawling /proc with that information, just more efficient and race-free.
> >> >> >>
> >> >> >> Doing it using the pid of writer is wrong. So is doing it with the
> >> >> >> cgroup of the writer. The fact that it's even possible to use the pid
> >> >> >> of the caller of write(2) is a mistake, but that particular mistake
> >> >> >> is, unfortunately, well-enshrined in history.
> >> >> >>
> >> >> >> >
> >> >> >> > I repeat, it is *not* access control.
> >> >> >> >
> >> >> >>
> >> >> >> Sure it is.
> >> >> >>
> >> >> >> Either correct attribution of logs doesn't matter, in which case it
> >> >> >> makes little difference how you do it, or it does matter, in which
> >> >> >> case it should be done right.
> >> >> >
> >> >> > Well journald can *also* get SO_PEERCGROUP and log anomalies if the 2
> >> >> > differ. That is if the log happens on a connected socket.
> >> >> >
> >> >> > If the log happens on a unix datagram* then SO_PEERCGROUP is not
> >> >> > available because there is no connect(), however write() cannot be used
> >> >> > either, only sendmsg() AFAIK, so the "setuid" binary attack does not
> >> >> > apply.
> >> >> >
> >> >>
> >> >> Or you could only send SCM_CGROUP when the writer asks sendmsg to send
> >> >> it, in which case this whole problem goes away.
> >> >
> >> > Sending SCM_CGROUP explicitly is also sending cgroup info at write(2) time
> >> > and if receiver uses that info for access control, it can be problematic.
> >> >
> >>
> >> Not really. write(2) can't send SCM_CGROUP. Callers of sendmsg(2)
> >> who supply SCM_CGROUP are explicitly indicating that they want their
> >> cgroup associated with that message. Callers of write(2) and send(2)
> >> are simply indicating that they have some bytes that they want to
> >> shove into whatever's at the other end of the fd.
> >
> > So you are telling me that you want to change all code that writes to
> > stderr to be changed to use sendmsg() instead of write() in order to get
> > that information ?
>
> No. I'm telling you that I want whoever writes the logging code to
> change *the logging code* to use sendmsg.
>
> > If you are using datagram sockets then the sender explicitly has to use
> > sendmsg() already and if a setuid binary can be convinced to send
> > arbitrary data to an arbitrary datagram sokcet you have bigger problems
> > in that binary, and said binary will send you whatever cgroup it is in.
>
> Really?
I want to retract my comment in light of the fact you can use connect()
on a datagram socket and then kernel will "helpfully" then allow you to
use write() on it, I forgot about this detail.
For the logging q. above I wouldn't see any issue unless the journald
people were planning on using connect() themselves to pie things like
stderr over a datagram socket.
If that were the case, well then ...
OTOH, I still need to understand how you attack a setuid binary to fake
a cgroup, it's not like uig/gid information that is changed by the
simple fact of invoking the setuid binary, and this is a brand new
contract, so it is not unreasonable to put in the security
considerations that a setuid binary should take extracare to where they
emit stdout/stderr message should they decide to change their cgroup ...
Simo.
next prev parent reply other threads:[~2014-04-17 18:31 UTC|newest]
Thread overview: 91+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-04-15 21:15 [PATCH 0/2] net: Implement SO_PEERCGROUP and SO_PASSCGROUP socket options Vivek Goyal
2014-04-15 21:15 ` [PATCH 1/2] net: Implement SO_PEERCGROUP Vivek Goyal
[not found] ` <1397596546-10153-2-git-send-email-vgoyal-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-15 21:54 ` Andy Lutomirski
2014-04-16 0:22 ` Vivek Goyal
2014-04-15 21:15 ` [PATCH 2/2] net: Implement SO_PASSCGROUP to enable passing cgroup path Vivek Goyal
[not found] ` <1397596546-10153-3-git-send-email-vgoyal-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-15 21:53 ` Andy Lutomirski
2014-04-15 23:09 ` Simo Sorce
2014-04-16 0:20 ` Vivek Goyal
[not found] ` <20140416002010.GA5035-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-16 1:05 ` David Miller
2014-04-16 3:47 ` Andy Lutomirski
[not found] ` <CALCETrWzHYN3kKcmDTFDfGhZqE4u9+6XDtiOu5nncbK_7KKH0g-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-16 10:17 ` Vivek Goyal
[not found] ` <20140416101709.GA14131-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-16 14:34 ` Andy Lutomirski
[not found] ` <CALCETrUTMSpd=NYn9QuO5Y3WY0uBhjNEHO0jCwZu0L59CpeDew-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-16 15:10 ` Vivek Goyal
2014-04-16 12:57 ` David Miller
2014-04-16 14:37 ` Andy Lutomirski
[not found] ` <CALCETrVv8SPM5xjOVGy7qO2aq3FKtG2uG57J49nO7Wy0-gg0Ew-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-16 16:13 ` Simo Sorce
2014-04-16 16:21 ` Tejun Heo
[not found] ` <20140416162149.GI1257-Gd/HAXX7CRxy/B6EtB590w@public.gmane.org>
2014-04-16 16:54 ` Simo Sorce
[not found] ` <1397664837.19767.410.camel-Hs+ccMQdwurzDu64bZtGtWD2FQJk+8+b@public.gmane.org>
2014-04-16 16:31 ` Andy Lutomirski
[not found] ` <CALCETrXn7b6UuALpGUVoyQYfR2uzk5tj2ABV=dkvtFNgqM5sxQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-16 17:02 ` Simo Sorce
2014-04-16 17:29 ` Andy Lutomirski
[not found] ` <CALCETrU_yKQVZyVug25cxwQFjWJ7Zf20FY-6ht+RJifXtDdDWg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-16 17:34 ` Simo Sorce
[not found] ` <1397669685.19767.450.camel-Hs+ccMQdwurzDu64bZtGtWD2FQJk+8+b@public.gmane.org>
2014-04-16 17:53 ` Andy Lutomirski
2014-04-16 18:36 ` Vivek Goyal
[not found] ` <20140416183614.GH31074-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-16 18:40 ` Andy Lutomirski
2014-04-16 18:51 ` Vivek Goyal
2014-04-16 18:59 ` Andy Lutomirski
2014-04-16 18:06 ` Vivek Goyal
[not found] ` <20140416180642.GG31074-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-16 18:13 ` Andy Lutomirski
2014-04-16 18:25 ` Vivek Goyal
[not found] ` <20140416182530.GB550-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-16 18:35 ` Andy Lutomirski
[not found] ` <CALCETrUs1js3Br81ZkiQnsuWduzOiqDe3aV0K_z_zw0znSuiag-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-16 19:06 ` Vivek Goyal
2014-04-16 19:13 ` Andy Lutomirski
[not found] ` <CALCETrUv56awd+UoO_f8LLL2FVq-Hc6Bd6iBGMqWjVGpgxgTSg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-16 19:39 ` Vivek Goyal
2014-04-16 20:24 ` Andy Lutomirski
2014-04-17 13:41 ` Vivek Goyal
[not found] ` <CALCETrVUw5+vCCONy1VTXpskbY_eZFo2CtbehwV5Mhj4d4+icw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-16 18:59 ` Vivek Goyal
2014-04-17 15:41 ` Daniel J Walsh
2014-04-17 16:05 ` Andy Lutomirski
[not found] ` <534FF61B.4010901-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-17 16:04 ` Simo Sorce
[not found] ` <1397750674.2628.44.camel-Hs+ccMQdwurzDu64bZtGtWD2FQJk+8+b@public.gmane.org>
2014-04-17 16:11 ` Andy Lutomirski
[not found] ` <CALCETrUrj2LtAoXp600BD2ANgE2UUEbTYQrK8hHqDR=qpeFPcg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-17 16:24 ` Simo Sorce
[not found] ` <1397751853.2628.50.camel-Hs+ccMQdwurzDu64bZtGtWD2FQJk+8+b@public.gmane.org>
2014-04-17 16:37 ` Andy Lutomirski
[not found] ` <CALCETrWNWvumFg9Qba0GEOjYop4WYe530tCPtakrhnoCrHqhUQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-17 16:48 ` Simo Sorce
[not found] ` <1397753323.2628.60.camel-Hs+ccMQdwurzDu64bZtGtWD2FQJk+8+b@public.gmane.org>
2014-04-17 16:55 ` Andy Lutomirski
[not found] ` <CALCETrXj6kD3E+vsaWmkrSbaQYTu=c-Hsw640jh4O+FbojYk2g-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-17 17:12 ` Vivek Goyal
2014-04-17 17:26 ` Andy Lutomirski
[not found] ` <CALCETrVq4HRpfWOAbZAQbyjuraQd=OxnW=WjSoe5JgBzRStiKg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-17 17:33 ` Simo Sorce
[not found] ` <1397756025.2628.64.camel-Hs+ccMQdwurzDu64bZtGtWD2FQJk+8+b@public.gmane.org>
2014-04-17 17:35 ` Andy Lutomirski
[not found] ` <CALCETrVBJFgKwRKBE2jAG6kiGgkJ+MyQiw2nyz5yj0h68kCk9A-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-17 17:47 ` Simo Sorce
[not found] ` <1397756821.2628.69.camel-Hs+ccMQdwurzDu64bZtGtWD2FQJk+8+b@public.gmane.org>
2014-04-17 18:05 ` Andy Lutomirski
2014-04-17 18:23 ` Simo Sorce
[not found] ` <1397759013.2628.86.camel-Hs+ccMQdwurzDu64bZtGtWD2FQJk+8+b@public.gmane.org>
2014-04-17 18:33 ` Andy Lutomirski
2014-04-17 18:50 ` Vivek Goyal
[not found] ` <20140417185023.GA32527-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-17 18:57 ` Vivek Goyal
[not found] ` <20140417185742.GB32527-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-17 19:06 ` Andy Lutomirski
[not found] ` <CALCETrXJPJeGBdauQS_WR5FNaZXR=05NjNKuC6r0xFORt+eaJQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-17 19:15 ` Simo Sorce
2014-04-17 19:19 ` Andy Lutomirski
2014-04-17 19:10 ` Simo Sorce
[not found] ` <1397761817.2628.113.camel-Hs+ccMQdwurzDu64bZtGtWD2FQJk+8+b@public.gmane.org>
2014-04-17 19:16 ` Vivek Goyal
[not found] ` <20140417191646.GA2461-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-17 19:46 ` Andy Lutomirski
[not found] ` <CALCETrW3F1+3qF3thrAmuoWVbRveBJ2=owpigh4mv6iAafoQCw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-21 15:03 ` Vivek Goyal
2014-04-21 15:47 ` Andy Lutomirski
2014-04-23 15:07 ` Vivek Goyal
2014-04-23 15:37 ` Andy Lutomirski
2014-04-23 16:01 ` Vivek Goyal
2014-04-17 19:23 ` Andy Lutomirski
2014-04-17 17:52 ` Simo Sorce
[not found] ` <1397757169.2628.75.camel-Hs+ccMQdwurzDu64bZtGtWD2FQJk+8+b@public.gmane.org>
2014-04-17 18:04 ` Andy Lutomirski
[not found] ` <CALCETrUon7mZzp12th2bZ=cJyTjn8ePwg_VtPWL_bykjnnpKLw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-17 18:31 ` Simo Sorce [this message]
2014-04-17 16:38 ` Vivek Goyal
2014-04-17 16:12 ` Vivek Goyal
2014-04-23 16:45 ` Vivek Goyal
2014-04-23 17:29 ` David Miller
[not found] ` <20140423.132955.671992126955940387.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
2014-04-24 20:34 ` Vivek Goyal
[not found] ` <20140424203427.GC19091-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-24 20:48 ` David Miller
2014-04-24 21:04 ` Vivek Goyal
2014-04-24 21:11 ` David Miller
[not found] ` <20140424.171155.806959282091051918.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
2014-04-25 0:29 ` Simo Sorce
[not found] ` <1397596546-10153-1-git-send-email-vgoyal-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-22 20:05 ` [PATCH 0/2] net: Implement SO_PEERCGROUP and SO_PASSCGROUP socket options David Miller
2014-04-22 20:08 ` Andy Lutomirski
2014-04-22 20:29 ` David Miller
2014-04-22 20:31 ` Andy Lutomirski
2014-04-22 20:32 ` David Miller
[not found] ` <20140422.163251.1863774803211446171.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
2014-04-23 0:37 ` Andy Lutomirski
[not found] ` <CALCETrX_TCbKy-3W590wG3rq9O3Hzbqc_wu3EGg7PKn2NNsTpQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-04-23 19:05 ` Vivek Goyal
2014-04-23 20:53 ` Daniel J Walsh
[not found] ` <5358284B.7020706-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-24 13:01 ` Vivek Goyal
[not found] ` <20140422.160558.627080587952506099.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
2014-04-23 15:55 ` Vivek Goyal
[not found] ` <20140423155512.GA24651-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-04-23 16:16 ` Vivek Goyal
2014-04-23 17:21 ` Andy Lutomirski
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1397759484.2628.93.camel@willson.li.ssimo.org \
--to=ssorce-h+wxahxf7alqt0dzr+alfa@public$(echo .)gmane.org \
--cc=cgroups-u79uwXL29TY76Z2rM5mHXA@public$(echo .)gmane.org \
--cc=davem-fT/PcQaiUtIeIZ0/mPfg9Q@public$(echo .)gmane.org \
--cc=dwalsh-H+wXaHxf7aLQT0dZR+AlfA@public$(echo .)gmane.org \
--cc=kay-H+wXaHxf7aLQT0dZR+AlfA@public$(echo .)gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public$(echo .)gmane.org \
--cc=lpoetter-H+wXaHxf7aLQT0dZR+AlfA@public$(echo .)gmane.org \
--cc=luto-kltTT9wpgjJwATOyAt5JVQ@public$(echo .)gmane.org \
--cc=netdev-u79uwXL29TY76Z2rM5mHXA@public$(echo .)gmane.org \
--cc=tj-DgEjT+Ai2ygdnm+yROfE0A@public$(echo .)gmane.org \
--cc=vgoyal-H+wXaHxf7aLQT0dZR+AlfA@public$(echo .)gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox