On Wed, 2008-11-05 at 11:38 -0500, Paul Moore wrote:
> On Wednesday 05 November 2008 9:34:42 am Eric Paris wrote:
> > Currently when SELinux has not been updated to handle a netlink
> > message type the operation is denied with EINVAL. This patch will
> > leave the audit/warning message so things get fixed but if policy
> > chose to allow unknowns this will allow the netlink operation.
> >
> > Signed-off-by: Eric Paris <eparis@redhat.com>
> > ---
> >
> > security/selinux/hooks.c | 2 +-
> > 1 files changed, 1 insertions(+), 1 deletions(-)
> >
> >
> > diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
> > index f85597a..c6f8f3e 100644
> > --- a/security/selinux/hooks.c
> > +++ b/security/selinux/hooks.c
> > @@ -4387,7 +4387,7 @@ static int selinux_nlmsg_perm(struct sock *sk,
> > struct sk_buff *skb) "SELinux: unrecognized netlink message"
> > " type=%hu for sclass=%hu\n",
> > nlh->nlmsg_type, isec->sclass);
> > - if (!selinux_enforcing)
> > + if (!selinux_enforcing || security_get_allow_unknown())
> > err = 0;
> > }
>
> What about moving the security_get_allow_unknown() call to the default
> switch clause of selinux_nlmsg_lookup()? Something like this:
>
> /* No messaging from userspace, or class unknown/unhandled */
> default:
> if (!security_get_allow_unknown())
> err = -ENOENT;
> break;
>
> This seems like a more natural fit to me (although maybe the audit
> message should be moved to selinux_nlmsg_lookup() too?) and it has the
> benefit of still checking the socket permissions via socket_has_perm()
> in the event that the netlink message is unknown.

We already just blindly allow the case where a new/unknown sclass is used which is what this part of the switch statement hits. I wanted to get the case where a known class has a new mesg type (aka nlmsg_perm returns -EINVAL)

Not sure that the socket check is worth anything since I don't (in either case) know what perms to ask for.

I also considered making the case of unknown msg type return ALL of the perms for that entire socket class but I think what I did is the best/easiest way we can go....

-Eric

--
This message was distributed to subscribers of the selinux mailing list.
If you no longer wish to subscribe, send mail to majordomo@tycho.nsa.gov with
the words "unsubscribe selinux" without quotes as the message.