Fixing CAP_SETPCAP

Linux capabilities have been around for almost ten years now – they were originally merged into a 2.1 kernel – but they haven't gotten a lot of use in that time. One pretty basic missing feature, support for associating capabilities with files, has been merged for 2.6.24. This allows a longstanding hack, which redefines the proper usage of CAP_SETPCAP, to be fixed; this too has been merged into 2.6.24.

A bit of review is probably in order. Capabilities are a way to separate individual privileges that are normally all granted to the root user. There are currently 31 different capabilities defined (in linux/capability.h), but there are efforts underway to allow for expansion. The idea is that a program should be able to set the system time, for example, without needing the entire set of privileges that come with a setuid(0) program.

Capabilities originally came from a proposed POSIX standard that was eventually not adopted, but, in the meantime, got included into Linux. The feature has languished since, for a number of reasons, but perhaps the largest was that there was no way to associate executable programs with a set of capability bits. Now that capability bits can be stored in the extended attributes of files, the process can get the proper capabilities when the program is invoked. Standard UNIX permissions still apply – users can only execute programs they have an x bit for.

In order to use capabilities at all, prior to being able to store them with files, a method was needed to set the capabilities of a running process. The CAP_SETPCAP capability was co-opted for this purpose. A process with this capability, which, in practice, meant root processes could set the capabilities of another process. If that process was meant to be able to do the same – something that needs to be carefully considered – it could get the CAP_SETPCAP bit as well.

This could really only be used to add capabilities to long running processes that were not run as root (which has all of the capabilities), or to remove some capabilities from daemons run as root. Other schemes using setuid wrappers for utility programs that needed some privileges could also be imagined, but distributions or tools that use capabilities are not widespread.

CAP_SETPCAP was never meant to have this behavior, so the recent patch restores it its original meaning. As odd as it might seem at first, CAP_SETPCAP is only meant to allow changes to a process's own capabilities; in fact, with this patch applied, there is no way for a process to change a running process's capabilities. That is probably the biggest user-visible change.

Capabilities are not a single set of bits, but are instead, three sets of bits representing the effective, permitted, and inheritable capabilities of a process. Files, similarly, have three capability sets which are combined with those of the process executing the file using the "capability rules" (described in the patch and in an LWN article from a year ago) to determine the three sets for the process created.

For processes, the effective set contains those capabilities currently enabled – a process might drop some that it is allowed once it has performed the corresponding privileged operation – while the permitted set is a superset of the effective set, including all capabilities allowed to that process. The inheritable set are those that are passed on to a new program started by an exec() call, which is where the new CAP_SETPCAP comes into play; a process with this capability can change its inheritable set to include any capability, including those that are not in their permitted set.

This allows processes to bestow privileges that they do not possess upon their children, which provides for some interesting uses. It helps further partition privileges by not requiring a process to have a particular capability simply to pass it on to children. The example provided in the patch illustrates this nicely: the login program does not require many privileges, but through some policy mechanism (pam_cap for example) could allow certain users to have extra capabilities. Because the login process does not itself possess those extra capabilities, this could limit the damage an exploit of login could do.

It is unclear whether these recent additions to the capability feature set will result in more capability users. There is a lot of work in the kernel security space right now as kernel hackers and security folks try to come up with sensible security solutions for Linux. The complexity of SELinux, along with the fact that many administrators disable it rather than try to figure it out, seems to have the community casting about for other solutions. It is possible that capabilities might be a part of another solution, though its complexities are far from trivial. Though most of the major distributions have already made their security model choice, a capabilities-based distribution would be interesting to see; it might make a nice project for a smaller, up-and-coming, distribution to try.

Unfortunately, the word "capabilities" in the security world has two not-very-similar
meanings: "POSIX capabilities" and "regular capabilities".  The POSIX folks wanted to add some
of the features of a "capability" system, but by the time they were done, they ended up with
sets of bits that really were nothing like a real capability-based system.  Capability-based
systems are uncommon today, but they exist; EROS is a good example.

I don't understand why the inheritance bits make things much safer. If login doesn't have a
capability, but its children do, a crack of login can just run a program with the needed
capability. It's an extra hoop to jump through, but I don't see where it adds much.

On that note, I actually agree with the article that "a capabilities-based distribution would
be interesting to see" -- except with, you know, real capabilities :-).

Unix has them, after all -- fds are exactly capabilities.  (A pipe is a kernel-space ring
buffer with two facets, etc.)  The model is a little inelegant because you have normal
numbered fds, the quasi-fds for "current directory" and "root directory" (all the syscalls
that take filenames are effectively calls against the interface these fds provides), and the
vast number of syscalls that are available to every process provides a little more exposure
than one would like... (kill(2), for instance, is annoying).  But these seem mostly solveable,
and in these days of bind mounts, FUSE, containers, etc., one could do pretty credible
POLA/capability-style containment with the stock Linux kernel and a custom userspace.

Plus it seems worthwhile to try, because capabilities are the best chance for surviving worms
and all the other wonderful conveniences of the modern internet, but so long as they only live
in pure-research OSes they aren't going anywhere.  Building on Linux gives you its existing
userspace to use as a base, and lets you move quickly to dealing with the real problems of
building a practical system that one can deploy, administer, etc.  (Eros is very cool, but how
the heck could one ever practically administer or even understand a system whose whole state
is a blob of gigabytes of memory tied together willy-nilly by pointers?)

Err... ObOnTopic: Building a system like I describe is much easier given the existence of
CAP_SYS_CHROOT.  (Though another option would be to eliminate the root dir entirely by
chrooting everything to a designated unreadable/unwriteable/empty directory, and just using
openat() etc all the time.  ...Too bad there's no execat().)

NJS:

Your idea sounds like Adam Langley's master's thesis:

http://www.imperialviolet.org/binary/pucs.pdf

The general use of CAP_SYS_ADMIN all around the kernel is what makes me doubt of any new
capability system extensions to what exists now.

I'm no security expert, but can't understand how they can overcame that without a major
overhaul all around the kernel code.

I see this new features as nice things to help secure a system, but this all seem just more
"hacks" around the fact there isn't any "grand scheme" for security well thought (I mean, in a
modern way, because we all know traditional UNIX security way is just obsolete).