Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 11 Jun 2015 17:50:32 +0000
From:      bugzilla-noreply@freebsd.org
To:        freebsd-pf@FreeBSD.org
Subject:   [Bug 200330] panic: pf_addr_cmp: unknown address family 0 when scrub fragment drop-ovl is used
Message-ID:  <bug-200330-17777-ZVb2oJ6QiZ@https.bugs.freebsd.org/bugzilla/>
In-Reply-To: <bug-200330-17777@https.bugs.freebsd.org/bugzilla/>
References:  <bug-200330-17777@https.bugs.freebsd.org/bugzilla/>

next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=200330

--- Comment #15 from Kristof Provost <kristof@freebsd.org> ---
The problem I have with 'drop-ovl' (and the same applies to 'crop') is that
it's a bit misleading.
They don't actually reassemble the fragmented packet. 

That means a far lower system load, but you can't actually filter in-depth. For
example, with 'scrub all fragment reassemble' you can do 'pass in inet proto
icmp all icmp-type echoreq keep state', but with 'scrub all fragment drop-ovl'
you'll end up just dropping a fragmented ICMP echo request. An unfragmented
ICMP echo request will just get through.
That goes against what I'd expect.

I think you (and really all pf users) want to be using 'scrub in all fragment
reassemble'.

(That said, if you can break it, please keep using drop-ovl, so I can debug it
;))

-- 
You are receiving this mail because:
You are on the CC list for the bug.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-200330-17777-ZVb2oJ6QiZ>