On 2010/09/20 11:10, Henning Brauer wrote:
>                                oh and when the snaplen is too small i
> don't do the rewrite and pass out the packet unmodified... not sure
> what else we could do but dropping, which would be inconsistent with
> other pcap stuff.

oh that's wierd indeed. would it be totally insane to print both the
natted and original addresses? then it would be clear that it's been
truncated.

> this touches the regular code patches (aka non-pflog) quite a bit too,
> there is some heavy code refactoring. watch you state tables
> carefully.

i haven't hit anything nasty yet.. (v4 and a little bit of v6).

Reply via email to