reassembled packets and pfil

[Luigi Rizzo]

On Tue, Apr 13, 2010 at 01:57:01PM +1200, Matthew Luckie wrote:
> Hi
> 
> Reassembled packets are not passed to the packet filter interface for
> both IPv4 and IPv6, so a firewall has no effect if the packets arrive
> in fragments.  Here is a patch to fix this for IPv6.  The patch for
> IPv4 is similarly trivial, but I have not written / tested it yet.
> 
> Is there any particular reason why reassembled packets were not
> checked?  If the answer is no, I'll send in a PR.

I think it was just a random decision -- either pass packets to
the firewall before reassembly as we do, or after reassembly, as
linux does. Both have pros and cons.
Going through the firewall twice, however, is problematic because
far too many things (counters, dummynet, etc.) expect to see each
packet only once.

I think that a patch like the one you propose is very useful (for
ipv4 as well) but it requires a sysctl or other mechanism to make
sure that when it is enabled we don't pass fragments through the
firewall.

cheers
luigi

> I've tested the patch below.
> 
> Matthew
> 
> --- sys/netinet6/frag6.c.orig   2008-11-25 15:59:29.000000000 +1300
> +++ sys/netinet6/frag6.c        2010-04-13 13:21:02.000000000 +1200
> @@ -46,6 +46,7 @@ __FBSDID("$FreeBSD: src/sys/netinet6/fra
>  
>  #include <net/if.h>
>  #include <net/route.h>
> +#include <net/pfil.h>
>  
>  #include <netinet/in.h>
>  #include <netinet/in_var.h>
> @@ -568,6 +569,13 @@ insert:
>         *offp = offset;
>  
>         IP6Q_UNLOCK();
> +
> +       if (PFIL_HOOKED(&inet6_pfil_hook) &&
> +               (pfil_run_hooks(&inet6_pfil_hook, &m, m->m_pkthdr.rcvif, PFIL_IN, NULL) ||
> +               m == NULL)) {
> +               return IPPROTO_DONE;
> +       }
> +
>         return nxt;
>  
>   dropfrag:
> _______________________________________________
> freebsd-net at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"