sandboxed-api/sandboxed_api/sandbox2/docs/faq.md
Christian Blichmann 177b969e8c
Sandboxed API OSS release.
PiperOrigin-RevId: 238996664
Change-Id: I9646527e2be68ee0b6b371572b7aafe967102e57

Signed-off-by: Christian Blichmann <cblichmann@google.com>
2019-03-18 19:00:48 +01:00

5.4 KiB

FAQ

Can I use threads?

Yes, threads are supported in sandbox2.

All threads must be sandboxed

Because of the way Linux works, the seccomp-bpf policy is applied to the current thread only: this means other existing threads do not get the policy, but future threads will inherit the policy.

If you are using sandbox2 in the default mode where sandboxing is enabled before execve(), all threads will inherit the policy, and there is no problem. This is the preferred mode of sandboxing.

If you want to use the second mode where the executor has set_enable_sandbox_before_exec(false) and the sandboxee tells the executor when it wants to be sandboxed with SandboxMeHere(), then the filter still needs to be applied to all threads. Otherwise, there is a risk of a sandbox escape: malicious code could migrate from a sandboxed thread to an unsandboxed thread.

The Linux kernel introduced the TSYNC flag in version 3.17, which allows applying a policy to all threads. Before this flag, it was only possible to apply the policy on a thread-by-thread basis.

If sandbox2 detects that it is running on a kernel without TSYNC-support and you call SandboxMeHere() from multi-threaded program, sandbox2 will abort, since this would compromise the safety of the sandbox.

How should I compile my sandboxee?

If not careful, it is easy to inherit a lot of dependencies and side effects (extra syscalls, file accesses or even network connections) which make sandboxing harder (tracking down all side effects) and less safe (because the syscall and file policies are wider). Some compile options can help reduce this:

  • statically compile the sandboxee binary to avoid dynamic linking which uses a lot of syscalls (open()/openat(), mmap(), etc.). Also since Bazel adds pie by default but static is incompatible with it, use the features flag to force it off. That is, use the following options in cc_binary rules:

    linkstatic = 1,
    features = [
      "fully_static_link",  # link libc statically
      "-pie",
    ],
    

    However: this has the downside of reducing ASLR heap entropy (from 30 bits to 8 bits), making exploits easier. Decide carefully what is preferable depending on your sandbox implementation and policy:

    • not static: good heap ASLR, potentially harder to get initial code execution but at the cost of a less effective sandbox policy, potentially easier to break out of.
    • static: bad heap ASLR, potentially easier to get initial code execution but a more effective sandbox policy, potentially harder to break out of.

    It is an unfortunate choice to make because the compiler does not support static PIE (Position Independent Executables). PIE is implemented by having the binary be a dynamic object, and the dynamic loader maps it at a random location before executing it. Then because the heap is traditionnally placed at a random offset after the base address of the binary (and expanded with brk syscall), it means for static binaries the heap ASLR entropy is only this offset because there is no PIE.

For examples of these compiling options, look at the static example BUILD.bazel: static_bin.cc is compiled statically, which allows us to have a very tight syscall policy. This works nicely for sandboxing third party binaries too.

Can I sandbox 32-bit x86 binaries?

Sandbox2 can only sandbox the same arch as it was compiled with.

In addition, support for 32-bit x86 has been removed from Sandbox2. If you try to use a 64-bit x86 executor to sandbox a 32-bit x86 binary, or a 64-bit x86 binary making 32-bit syscalls (via int 0x80), both will generate a sandbox violation that can be identified with the architecture label [X86-32].

The reason behind this behavior is that syscall numbers are different between architectures and since the syscall policy is written in the architecture of the executor, it would be dangerous to allow a different architecture for the sandboxee. Indeed, allowing an seemingly harmless syscall that in fact means another more harmful syscall could open up the sandbox to an escape.

Any limits on the number of sandboxes an executor process can request?

For each sandboxee instance (new process spawned from the forkserver) a new thread is created - that's where the limitation would lie.

Can an Executor request the creation of more than one Sandbox?

No. There is a 1:1 correspondence - an Executor instance stores the PID of the sandboxee, manages the Comms instance to the Sandbox instance, etc.

Can I use sandbox2 from Go?

Yes. Write your executor in C++ and expose it to Go via SWIG.

Why do I get Function not implemented inside forkserver.cc?

Sandbox2 only supports running on reasonably new kernels. Our current cut-off is the 3.19 kernel though that might change in the future. The reason for this is that we are using relatively new kernel features including user namespaces and seccomp with the TSYNC flag.

If you are running on prod, this should not be in issue, since almost the entire fleet is running a new enough kernel. If you have any issues with this, please contact us.

If you are running on Debian or Ubuntu, updating your kernel is as easy as apt-get install linux-image-[recent version].