On Wed, Feb 25, 2009 at 5:15 PM, Jason Dixon <ja...@dixongroup.net> wrote:
> On Wed, Feb 25, 2009 at 04:39:08PM -0800, patrick keshishian wrote:
>> On Wed, Feb 25, 2009 at 4:07 PM, ropers <rop...@gmail.com> wrote:
>> > 2009/2/25 patrick keshishian <pkesh...@gmail.com>:
>> >> On Tue, Feb 24, 2009 at 9:48 PM, Hilco Wijbenga
>> >> <hilco.wijbe...@gmail.com> wrote:
>> >>> 2009/2/23 Jason Dixon <ja...@dixongroup.net>:
>> >>>> ##########################################################
>> >>>> 00 ext_if = "sk0"
>> >>>> 01 int_if = "sk1"
>> >>>> 02
>> >>>> 03 set skip on lo
>> >>>> 04
>> >>>> 05 scrub in
>> >>>> 06
>> >>>> 07 nat on $ext_if from $int_if:network to any -> ($ext_if:0)
>> >>>> 08
>> >>>> 09 block in log all
>> >>>> 10 pass in on $int_if inet keep state
>> >>
>> >> # I think you are missing a pass out on $ext_if rule
>> >> 11 pass out on $ext_if
>> >>
>> >> w/o 11 all inbound packets are blocked by 09.
>> >
>> > Please correct me it I'm wrong, but my understanding is that Jason's
>> > line 09 is only blocking packets that traverse interfaces in the
>> > *inward* direction (i.e. from the attached networks (WAN and LAN)
>> > *in*to the OpenBSD box). And I think that's why a "pass out" rule
>> > would be unnecessary, as the default is to pass packets. Again, take
>> > this with truckloads of salt; I could be wrong, but I trust Jason.
>>
>> You need states created for traffic passing through the pf firewall,
>> specifically through the $ext_if to allow packets flowing back in,
>> otherwise line 09 blocks those packets. I don't see where states would
>> get created for outbound traffic with Jason's rule-set.
>
> Using floating states (the default), line 10 should suffice. B But I
> forgot to account for outbound DNS requests from the firewall itself.
> Here is the same ruleset with the one extra rule. B It shouldn't be
> necessary for outbound traffic originating from $int_if:network, but it
> won't hurt either.
The floating states based on line 10 would be for pre-NAT sources on
$int_if and wouldn't match any inbound packets on $ext_if. Unless I'm
misunderstanding how NAT works with pf, there are no pass out rules
that would create states for these packets:
from pf.conf(5):
Since translation occurs before filtering the filter engine will see
packets as they look after any addresses and ports have been translated.
Filter rules will therefore have to filter based on the translated ad-
dress and port number. Packets that match a translation rule are only
automatically passed if the pass modifier is given, otherwise they are
still subject to block and pass rules.
...
Translation rules apply only to packets that pass through the specified
interface, and if no interface is specified, translation is applied to
packets on all interfaces.
--patrick
