dhcpleased (and a few other daemons) use bpf, thus see raw packets
from the wire before pf can block them. Most daemons of this type
also use bpf to send packets, and pf doesn't see these either.
This behaviour is intentional, and useful.
beebeet...@posteo.de wrote:
> Hi all,
>
> I'm running OpenBSD 6.9 as a home router, and observed some behavior of
> pf that I can't really make sense of.
>
> The router runs dhcpleased to obtain its IP address from the ISP, and
> I have
> the following pf rules (only the relevant ones are shown):
>
> block drop all
> pass out on $ext_if inet from ($ext_if) to ! <ipv4_reserved>
> block drop in log (all) on $ext_if inet proto udp from port 67 to port
> 68
> pass in on $ext_if inet proto udp from port 67 to 255.255.255.255 port
> 68
>
> (I need to mention here that after digging into some old discussions
> on the
> mailing list, I realize that the last two rules are unnecessary
> because DHCP
> traffic is supposedly processed by dhcpleased though bpf regardless of
> pf's
> decision, but my question is something else)
>
> With tcpdump on the external interface, I see packets similar to the
> following
> for lease renewal:
>
> {timestamp} {my_ip}.68 > {ip1}.67: xid:0xfe51c9a3 [|bootp]
> {timestamp} {ip2}.67 > {my_ip}.68: xid:0xfe51c9a3 Y:{my_ip} G:{ip1}
> [|bootp]
>
> Note that DHCP renew request is sent to {ip1}, but the DHCP
> acknowledgment
> is from {ip2}, so I guess {ip1} is a DHCP relay?
>
> The problems is, with my existing pf rules I expect the second packet
> to be
> blocked and logged to pflog0, but in reality, a tcpdump on pflog0
> shows that no
> packets are being blocked:
> tcpdump -l -n -i pflog0
>
> Why is the second packet not blocked by pf when its source ip address
> {ip2} is
> supposedly not in the state table?
>
> I would greatly appreciate any help on this.
>
> Best Regards
>
