Re: [PATCH] seccomp: add ptrace commands for suspend/resume

[Tycho Andersen <tycho.andersen-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org>]

On Tue, Jun 02, 2015 at 12:36:16PM +0300, Andrey Wagin wrote:
> 2015-06-01 22:28 GMT+03:00 Tycho Andersen <tycho.andersen-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org>:
> > This patch is the first step in enabling checkpoint/restore of processes
> > with seccomp enabled.
> >
> > One of the things CRIU does while dumping tasks is inject code into them
> > via ptrace to collect information that is only available to the process
> > itself. However, if we are in a seccomp mode where these processes are
> > prohibited from making these syscalls, then what CRIU does kills the task.
> >
> > This patch adds a new ptrace command, PTRACE_SUSPEND_SECCOMP that enables a
> > task from the init user namespace which has CAP_SYS_ADMIN to disable (and
> > re-enable) seccomp filters for another task so that they can be
> > successfully dumped (and restored).
> 
> Do we need to re-enable seccomp if a tracer detaches unexpectedly.
> CRIU can be killed and we should try to not affect the task state even
> in this case.

Yes, I think Pavel's suggestion on the CRIU list of simply
automatically re-enabling seccomp on ptrace detach is the right way to
go here; it should cover this case. The only question is whether or
not to leave the explicit ability to re-enable seccomp before detach
or not. I don't think it's necessary for CRIU, so perhaps I'll remove
it in the next version.

Tycho