|
Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com |
Subject: Re: Linux capability bounding set weakness
From: Paul Wouters (paul
XTDNET.NL)Date: Tue Jun 27 2000 - 15:50:44 CDT
- Next message: Paul Starzetz: "Improved ARP sniffer"
- Previous message: debian-security-announce
LISTS.DEBIAN.ORG: "[SECURITY] New verion of dhcp released"
- In reply to: Patrick Reynolds: "Linux capability bounding set weakness"
- Next in thread: Matthew Kirkwood: "Re: Linux capability bounding set weakness"
- Reply: Paul Wouters: "Re: Linux capability bounding set weakness"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Mon, 26 Jun 2000, Patrick Reynolds wrote:
> To make capability bounding sets at all useful, you have to disable
> CAP_SYS_RAWIO, which governs access to /dev/mem. Be advised that doing so
> will break X and any other user-space program that needs raw access to
> memory or I/O ports.
> Fix: if you disable anything in the capability bounding set, you must also
> disable CAP_SYS_RAWIO and CAP_SYS_MODULE.
These issues have been address a long time ago with LIDS (www.lids.org).
There, not init, but a special program called lidsadm is the control center
of the capabilities. It has clear documentation on why you MUST restrict
certain capabilities, and even has the option to compile a hardcoded list of
processes (such as X :) that can access /dev/mem despite the capability
setting.
I can recommend lids as a VERY good way to secure your system so much, you'll
find it impossible to cleanyl shutdown or reboot altogehter :)
See http://www.ota.be/linux/workshops/20000527/ for a RealMedia overview
of LIDS that I gave a few weeks ago for the OTA.
Paul
-- Only the access to the source code of our future television sets will guarantee the independence of content and technology.--- www.linuxtv.org
- Next message: Paul Starzetz: "Improved ARP sniffer"
- Previous message: debian-security-announce
LISTS.DEBIAN.ORG: "[SECURITY] New verion of dhcp released"
- In reply to: Patrick Reynolds: "Linux capability bounding set weakness"
- Next in thread: Matthew Kirkwood: "Re: Linux capability bounding set weakness"
- Reply: Paul Wouters: "Re: Linux capability bounding set weakness"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]